Nid 00 fake Bye DD % ore, Meer ebaan sae Lad FRG bey MERA ADA NS na Aram Ps Fo AtIEOG D San hota ons oe Wrens 425 rs Sood Re ‘ ; SRM A BALE s 5 . * ‘ a ‘ ‘ se heat a: aegis ace : " * . on et fh My Me ob ave “ Y - eh: {EER TS NPT HE 3 ‘ : Fe ao stale Lesh Re eae : + \ . i eat ert j : aC Lickaby ayed tras ¢ : : . " y et + + anita se ote i Ls : rc bs : . : ween ane oe ro Aste he ean aor a ~ ae ces 4 Site | ‘ =a a: Apa enananeeas A Bonide RE ‘ % : y . fi ‘3 : ra. : ote nes - en rvs * , ook Anns Mae APY Nayctaneetoeamoncn one 4 : . ‘: x é. ow Sessa parecer eer oy oe od Saas fax mate AaB¥AoI yy ote h hd se § eee SEE sooty rrbead Co re eee 2a amet ot Wea bye rae fein yA Pra Pe M3 Ae PIAL Seas ey Sy . eohstng Dig SARE ees PE Oe Re y ~ Poe eigeiseeness ora” f gt, eel del hee Pe fk : : J is i; i od sree : ‘ J , > hi . 3 jee % td de ae ee eae 1 nage 9 Sosa co yee oe : Vee ; by i ; : Sandbyieahe det NE eae ™ nee Ease oeeg z es etree afd wis. ae aiid Beir pale ’ ; feat ats MY ed hey k ta ity Bs . - Pa pena pe: 5 “4 . . ¥ ‘ 7 \4 ? hit ? * Jaca 5 Ke pee ee Teal pete A a ’ slat Ee Ss 5 : cola eed At aes “ oot aes i nit phy - : xr? se ; wee x an eee - . 3 2 P \ : : : ‘ Seng ey ae 2 PASSES Sy rte t a Ae ens ae ae aie ik : oO papeae 4m: 2 ‘ i : ? ivi sare = Pegs tse # Se 2e* : : 4 : ; “i : z dee ‘ ; iere tsb se? Want xp aes et t ; ; ewe : Bony em : 7 J : “4 4 24 oS Eo sgn [Res deb ae eA SYS em FTE? 3 re : San ‘3 x Pee ceanive oe 2) KU SU Ae otras fof at eae aa Seok co on dinetyesane ae Bu ee Geese bas pe tae (anesays eta e ie oe er oe eo ed wee to Cpayate. aes Vases ls God seens sos stes Ses gad- uh e-4aee Se; e Sane aes a eg sate ‘. ‘seatelSatetar er ahe Le Wee pet an Stee ASE ome eee es hagas Sap FE ALS Ne at aaa ae wee a i Bey: s ‘pp DOr Lae ay aa aa . eA ALE a ah ota 1a, x a. pops ea . : ; vets de ‘ ‘ ; é : “ Sa c wy ‘ ent T fm np tt Fatwa ‘ rts es : : ia a we SS eae ‘fe ~. ‘ > 4 F) ad aed (" Parr ence eho e eal anu mo Nowe A) ee ee er a, ee ” — wn — a) yc eh aoe ” | ee NOILNLILSNI NVINOSHLINS Sa 1yYVvuad nr BRARI ES SMITHSONIAN NOILAL = z = Re ea oe Z Ed 5 Zz = ae Sas ae Zz ar ac O el Ea DO Ne = oO oh nw ”n YW Ww Ny . Ww hie n O se Oo 3 a \. oO ue oe 2 FE Z = » 2 E Z y FA op) Ze. ” * Ze ” 2 q! 1_ LIBRARIES SMITHSONIAN _INSTITUTION NOILNLILSNI LIBRA Zz Hi Zz te = ” Z n Es ” i wo i ” = a S - fe = fae e. (es i ; Cc PS C S c zal fae} a m Et a = O or e) = ro) S Oo Fs - = I a : al Zz CIS isnt rr tlavea!T EIBRARIES SMITHSONIAN INSTITOTEOs NOILN _ 3 E Aah 3 = = ca = S = > = =) a =a) aa = om im > - », > = E é E es = hes = nm” w = ph fe D z B z = LIBRARIES SMITHSONIAN INSTITUTION NOILNLILSNI_NVINOSHLINS Saiuvugi7_LIBRA 2 INSTITUTION NOILAL NVINOSHLIWS SMITHSONIAN NVINOSHLIWS SMITHSONIAN f NVINOSHLIWS X YY oa. NOILNLILSNI_ NVINOSHLIWS Sa1uvyal LIBRARIES SJIMVNAIF LIBRARIES SMITHSONIAN LIBRARIES NOILMLILSNI LIBRARIES NOILNLILSNI NOILNLILSNI LIBRAR 1ES SMITHSONIAN 1 LIBRARIES SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLINS S3I1YVUGIT LIBRA Saluvugii saruvudit YN NOILALILSNI NVINOSHLINS S3IYVYalT LIBRARIES INSTITUTION INSTITUTION INSTITUTION SMITHSONIAN INSTITUTION NVINOSHLINS S3IYVUSEII NVINOSHLIWS SMITHSONIAN SMITHSONIAN GL) hb >” bk Ae pad NVINOSHLIWS NVINOSHLINS A TC SMITHSONIAN * : 1 LIBRARIES SMITHSONIAN INSTITUTION NOILOLILSNI NVINOSHLINS S3IYVYEIT_ LIBRA LIBRARIES NOILALILSNI NOILALILSNI NOILNLILSNI LIBRARIES SaiadVuYdit LIBRARTES N NOILNLILSNI SMITHSONIAN INSTITUTION NOILN- S3i1uYvVYaIT LIBRARIES INSTITUTION NOILONLILSNI = = ion ay = S S : 6 = = es = = = = 9) : 5 2 ie ze — = o ae i < 2 y = Be z a EB Zz WAS TLULIBRARIES SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLINS Saluvudi7_ LIBRA < = = = . g = = J y z RN = a > aa Ve fi > =~ Vi = \ = S Me Y flys bE eo SS = S Z 8 YY li 6 Mp fr = “X S34 ~ IVINOSHLINS S3IYVYE!I] LIBRARIES SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLI¥ INS Se NS Se. N‘ NVINOSHLIWS QL Ws YS ASN YS SMITHSONIAN NVINOSHLIWS SMITHSONIAN NVINOSHLIWS wae ASS SY SMITHSONIAN * MITHSONIAN INSTITUTION NOILONLILSNI LIBRARIES SMITHSONI/ By be ma i ate : : : 5 z es = 4 a = SA ee Ze 2 < = < : GM ac oc x« GYuy z. ro. a Vy : = = e = a Vif ae =f Pd > a Fs = ea lYVvug ae LIBRARI ES_ SMITHSONIAN INSTITUTION NOILALILSNI_NVINOSHLIV fe) — O ) = S amet a3 = Sy = = = = ee] = > = “D> 2 > = = — ‘2 = ae BAS 2 m 2 : m 2 m Le = op) nae ~ = (op) MITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLINS S31u¥WYuRgIT LIBRARIES a W z 72) ee 7) ee = x = a a = Fa = ne = = = oO a Ae ae oO a = >’ = > = ~ WS > “” Fs wn Fc ” - = LIBRARIES SMITHSONIAN INSTITUTION NOILONLILSNI NVINOSHLII! o ” . 0 =a tu a Ww 7 ud ” = “4 re a a = SAN al ioe | =i Ne = c = hae = ( as rae) a om. a rae) om coe = O = O ae O ce} =z ey) za =) Zz SMITHSONIAN INSTITUTION NOILONLILSNI NVINOSHLINS S3IYVYUE!IT LIBRARIES SMITHSONI saruvudi INSTITUTION Saiuvyadl Saleva dial INSTITUTION WWINOSHLINS S3IYVYGIT LIBRARIES SMITHSONIAN INSTITUTION NOILALILSNI NVINOSHLI INSTITUTION w Ze wn = 2) = = if a NS = sk ES aes 2) ar ae ANNE = x Bho SW Ft S = 6 BGs ERR 8 2 a5 2 z “fy = NS 2 = Zz = ia Te ae MITHSONIAN _INSTITUTION NOILNLILSNI NVINOSHLINS S3IYVYSGIT LIBRARIES SMITHSONI, — (ep) =z, Y +o e a Ww us b us = ax , o = a : = = t LPL Es ~ m = a = S = ” SMITHSONIAN INSTITUTION NOILALILSNI NVINOSHLIWS SaluYvygit THSONIAN JOSHLIWS THSONIAN NOSHLIWS THSONIAN NOSHLIWS 2g Ni mw. Gs ANNALS OF THE ANNALE VAN DIE SOUTH AFRICAN MUSEUM SUID-AFRIKAANSE MUSEUM VOLUME 86 BAND 86 ANNALS OF THE SOUTH AFRICAN MUSEUM ANNALE VAN DIE SUID-AFRIKAANSE MUSEUM VOLUME 86 BAND HE TRUSTEES OF THE DIE TRUSTEES VAN DIE SOUTH AFRICAN MUSEUM SUID-AFRIKAANSE MUSEUM GAPE TOWN KAAPSTAD 1981-1982 SET, PRINTED AND BOUND IN THE REPUBLIC OF SOUTH AFRICA BY THE RUSTICA PRESS (PTY.) LTD., WYNBERG, CAPE D54 List OF CONTENTS Page CurRiE, P. J. The osteology and relationships of Tangasaurus mennelli Haughton (Reptilia, osuchia) (Published: May 19825)).— sae save 44 oe cis ie Mes owen A eae 247 GosLINER, T. M. The South African Janolidae (Mollusca, Nudibranchia) with the description of a new genus and two new species. (Published September 1981.)................ 1 GRINE, F. E. Description of some juvenile hominid specimens from Swartkrans, Transvaal. CEnblisneGlOctober 198M.) i sxe Seok ee ce Mig ae ene mre Rp Tae Sin i Set 43 GRINE, F. E. Occlusal morphology of the mandibular permanent molars of the South African Negro and the Kalahari San (Bushman). (Published October 1981.) .......... IS y7/ GRInE, F. E. Relative sizes of the maxillary deciduous canine and central incisor teeth in the Kalahari San (Bushman) and South African Negro. (Published December ASUS) rea Rte ee baby trae alae a Sig there io ROS Eee) si fee Re ee ere 229 GRINE, F. E. see VAN DEN HEEVER, J. A. KENNEDY, W. J., KLINGER, H. C. & SUMMESBERGER, H. Cretaceous faunas from Zululand and Natal, South Africa. Additional observations on the ammonite subfamily Texanitinae Collignon, 1948. (Published October GIL) Ss OUNCE sean Rae elle a ate isk aed ee ee SCE SS, at, is ot OR 115 KLEIN, R. G. see Scott, L. KLINGER, H. C. see KENNEDY, W. J. Scorr, L. & KLEIN, R. G. A hyena-accumulated bone assemblage from Late Holocene deposits at Deelpan, @ranecibree state: (Published December 19812) 22... seh te gee eee te oe 217 SUMMESBERGER, H. see KENNEDY, W. J. VAN DEN HEEVER, J. A. & GRINE, F. E. Dinocephalia type material in the South African Museum (Reptilia, Therapsida). MiublishedvOctober lO Sila) ete aera cayenne Wea tee Cyr rm ete, Sua tus ee a tae ye 73 NEW GENERIC NAMES PROPOSED IN THIS VOLUME Bonisa Gosliner, 1981 SEPTEMBER 1981 ISSN 0303-2515 PA Mind 7 ‘OF THE SOUTH AFRICAN “MUSEUM CAPE TOWN INSTRUCTIONS TO AUTHORS lt. MATERIAL should be original and not published elsewhere, in whole or in part. 2. LAYOUT should be as follows: (a) Centred masthead to consist of Title: informative but concise, without abbreviations and not including the names of new genera or species Author’s(s’) name(s) Address(es) of author(s) (institution where work was carried out) Number of illustrations (figures, enumerated maps and tables, in this order) (b) Abstract of not more than 200 words, intelligible to the reader without reference to the text (c) Table of contents giving hierarchy of headings and subheadings (d) Introduction 2 (e) Subject-matter of the paper, divided into sections to correspond with those given in table of contents (f) Summary, if paper is lengthy (g) Acknowledgements (h) References (i) Abbreviations, where these are numerous 3. MANUSCRIPT, to be submitted in triplicate, should be typewritten and neat, double spaced with 2,5 cm margins all round. First lines of paragraphs should be indented. Tables and a list of legends for illustrations should be typed separately, their positions indicated in the text. All pages should be numbered consecutively. Major headings of the paper are centred capitals; first subheadings are shouldered small capitals; second subheadings are shouldered italics; third subheadings are indented, shouldered italics. Further subdivisions should be avoided, as also enumeration (never roman numerals) of headings and abbreviations. Footnotes should be avoided unless they are short and essential. Only generic and specific names should be underlined to indicate italics; all other marking up should be left to editor and publisher. 4. ILLUSTRATIONS should be reducible to a size not exceeding 12 « 18 cm (19 cm including legend); the reduction or enlargement required should be indicated; originals larger than 35 x 47 cm should not be submitted; photographs should be rectangular in shape and final size. A metric scale should appear with all illustrations, otherwise magnification or reduction should be given in the legend; if the latter, then the final reduction or enlargement should be taken into consideration. All illustrations, whether line drawings or photographs, should be termed figures (plates are not printed; half-tones will appear in their proper place in the text) and numbered in a single series. Items of composite figures should be designated by capital letters; lettering of figures is not set in type and should be in lower-case letters. The number of the figure should be lightly marked in pencil on the back of each illustration. 5. REFERENCES cited in text and synonymies should all be included in the list at the end of the paper, using the Harvard System (ibid., idem, loc. cit., op. cit. are not acceptable): (a) Author’s name and year of publication given in text, e.g.: ‘Smith (1969) describes .. .’ ‘Smith (1969: 36, fig. 16) describes...’ “As described (Smith 1969a, 19696; Jones 1971)’ ‘As described (Haughton & Broom 1927)...’ ‘As described (Haughton et al. 1927)...’ Note: no comma separating name and year Dagination indicated by colon, not p. names of joint authors connected by ampersand - et al. in text for more than two joint authors, but names of all authors given in list of references. (b) Full references at the end of the paper, arranged alphabetically by names, chronologically within each name, with suffixes a, b, etc. to the year for more than one paper by the same author in that year, e.g. Smith (1969a, 19695) and not Smith (1969, 1969a). For books give title in italics, edition, volume number, place of publication, publisher. For journal article give title of article, title of journal in italics (abbreviated according to the World list oa, scientific periodicals. 4th ed. London: Butterworths, 1963), series in parentheses, volume number, part number (only if independently paged) in parentheses, pagination (first and last pages of article). Examples (note capitalization and punctuation) BULLOUGH, W. S. 1960. Practical invertebrate anatomy. 2nd ed. London: Macmillan. FIsCHER, P.—H. 1948. Données sur la résistance et de le vitalité des mollusques. J. Conch., Paris 88: 100-140. Fiscuer, P.-H., DuvAL, M. & RaArFy, A. 1933. Etudes sur les échanges respiratoires des littorines. Archs Zool. exp. gén. 74: 627-634. Konn, A. J. 1960a. Ecological notes on Conus (Mollusca: Gastropoda) in the Trincomalee region of Ceylon. Ann. Mag. nat. Hist. (13) 2: 309-320. Konn, A. J. 19606. Spawning behaviour, egg masses and larval development in Conus from the Indian Ocean. Bull. Bingham oceanogr. Coll. 17 (4): 1-51. THELE, J. 1910. Mollusca: B. Polyplacophora, Gastropoda marina, Bivalvia. In: SCHULTZE, L. Zoologische und anthropologische Ergebnisse einer Forschungsreise im westlichen und zentralen Siid-Afrika 4: 269-270. Jena: Fischer. Denkschr. med.-naturw. Ges. Jena 16: 269-270. (continued inside back cover) bthes date nt Ae Lie fil ANNALS OF THE SOUTH AFRICAN MUSEUM ANNALE VAN DIE SUID-AFRIKAANSE MUSEUM Volume 86 Band September 1981 September Part 1 Deel THE SOUTH AFRICAN JANOLIDAE (MOLLUSCA, NUDIBRANCHIA) mare THE DESCRIPTION.OF A NEW GENUS AND TWO NEW SPECIES By TERRENCE M. GOSEINER Cape Town Kaapstad The ANNALS OF THE SOUTH AFRICAN MUSEUM are issued in parts at irregular intervals as material becomes available Obtainable from the South African Museum, P.O. Box 61, Cape Town 8000 Die ANNALE VAN DIE SUID-AFRIKAANSE MUSEUM word uitgegee in dele op ongereelde tye na gelang van die beskikbaarheid van stof Verkrygbaar van die Suid-Afrikaanse Museum, Posbus 61, Kaapstad 8000 OUT OF PRINT/UIT DRUK iy 20S, 5-8), SGD, 2S, 8, ico), SES, 5, 7-5), G(s ep ib), (=a) 8 OC 2a) O32) INGED, 5, 7, tor), 15S), WO), 2, BGS), 2X), 28, 250) EDITOR/REDAKTRISE Ione Rudner Copyright enquiries to the South African Museum Kopieregnavrae aan die Suid-Afrikaanse Museum ISBN 0 86813 022 2 Printed in South Africa by In Suid-Afrika gedruk deur The Rustica Press, Pty., Ltd., Die Rustica-pers, Edms., Bpk., Court Road, Wynberg, Cape Courtweg, Wynberg, Kaap THE SOUTH AFRICAN JANOLIDAE (MOLLUSCA, NUDIBRANCHIA) WITH THE DESCRIPTION OF A NEW GENUS AND TWO NEW SPECIES By TERRENCE M. GOSLINER South African Museum, Cape Town (With 28 figures and 3 tables) [MS accepted 28 April 1981] ABSTRACT Examination of the shallow waters of the Cape Peninsula, South Africa, has revealed the presence of three sympatric species of Janolidae. The description of Janolus capensis Bergh, 1907 is expanded. Janolus longidentatus sp. nov. is described in detail and, despite its resemblance to J. capensis, possesses several consistently distinct morphological characteristics. The two species differ in body shape, position of the gonopore and nephroproct, dentition of the jaws and radula, arrangement of ganglia within the central nervous system, and in several aspects of reproductive morphology. The third species differs significantly from all other members of the Janolidae in that the digestive gland surrounds the stomach rather than entering the notum and cerata. Bonisa nakaza gen. et sp. nov. is described in detail. The presence of intermediate morphological characters in several species of janolids necessitates the synonomy of Janolus Bergh, 1884, and Antiopella Hoyle, 1902. CONTENTS PAGE MNILOMUCHON js tine aes hee teen Ge ate oe 1 Janolus capensis Benen lO07. 5 generac. Aas ae. tan 2 NOCOTS UDGOIGHCHIS S05 Wl aeecbheccsesceaoncses 9 Bonisa nakaza Senn etispsnOVei. 44-25 os eek eee 19 DIScussionvOK ne anolidace sys oe coe eee 26 PNCKAMOWLE GS CMICINS ee hay ficients acca teredire ts Cuenta rae Rennes ot 40 INCIERENCES Men aie heii en ee aay en Nea se 41 INTRODUCTION The Janolidae are a family of arminacean nudibranchs consisting of 18 species in 5 genera. They are widely distributed in the temperate and tropical waters of the world. The genus Caldukia Burn & Miller, 1969, and the monotypic genus Galeojanolus Miller, 1971, are known only from Australia and New Zealand. Proctonotus Alder & Hancock, 1844, is also monotypic and has been reported only from Great Britain and the Brittany coast of France (Thompson & Brown 1976). Antiopella Hoyle, 1902, and Janolus Bergh, 1884, are much more widely distributed. Ann. S. Afr. Mus. 86 (1), 1981: 1-42, 28 figs, 3 tables. 2 ANNALS OF THE SOUTH AFRICAN MUSEUM Bergh (1907) described Janolus capensis from three specimens dredged from False Bay, south-western Cape. Intertidal and subtidal collections by means of S.C.U.B.A. around the Cape Peninsula have yielded specimens of Janolus capensis as well as two previously unknown species, which are here described. Antiopella and Janolus have been considered as distinct genera by most workers who have studied the Janolidae (Eliot 1906; Pruvot-Fol 1954; Marcus 1958; Burn & Miller 1969; Miller 1971). The South African species possess morphological features which are intermediate between those utilized to separ- ate Antiopella from Janolus. For this reason, a review of the generic status of the Janolidae is provided. To supplement this review, specimens of Janolus fuscus from California, J. cristatus from England, and J. toyamensis from _ Hawaii were also examined. Janolus capensis Bergh, 1907 Figs 1A-B, 2-9 Janolus capensis Bergh, 1907: 90, pl. 7 (figs 6-21). Barnard, 1927: 207, pl. 20 (figs 6-7). Material examined 10 specimens, 20 m depth off Llandudno (34°01’'S 18°20'E), 23 December ORE 6 specimens, intertidal, Clovelly, False Bay (34°05’S 18°26’E), 17-18 January 1980 3 specimens, 10 m depth, Rooi Els (34°18’S 18°49’E), 23 January 1980 1 specimen, 10 m depth, Castle Rocks, False Bay (34°18’S 18°29’E), 12 February 1980 5 specimens, 10 m depth, Windmill Beach, False Bay (34°12'S 18°27’E), 9 October 1980 2 specimens, 10 m depth, Rooi Els (34°18’S 18°49’E), 26 October 1980 2 specimens, 5 m depth, Miller’s Point, False Bay (34°14’S 18°29’E), 28 October 1980 Distribution Atlantic and Indian Ocean coasts of the Cape Peninsula, extending east- ward to Rooi Els, Cape Hangklip. External morphology The living animals (Fig. 1A—B) attain a length of 30 mm. The stout body is broadest anteriorly, tapering to the acute posterior end of the foot. The rhinophores (Fig. 2A) are perfoliate with 11 to 16 complete or incomplete transverse lamellae. Between the rhinophores is the large, convoluted inter- rhinophoral crest (caruncle) (Fig. 2B). The anus in middorsal, near the poster- ior limit of the notum. The head (Fig. 2C) is rounded with the mouth situated SOUTH AFRICAN JANOLIDAE 3 Fig. 2. Janolus capensis Bergh, 1907. A. Rhinophore, scale 1,0 mm. B. Inter-rhinophoral crest, scale 1,0 mm. C. Ventral view of head, scale 2,0 mm. D. Lateral view, scale 3,0 mm. Figs 2A-B, D are drawn from preserved material, Fig. 2C is from living material. centrally. Short, blunt dorsoventrally flattened oral tentacles are present on either side of the head. The foot is rounded anteriorly with a deep transverse groove. The separate male and female gonopores are situated on the right side near the middle of the body, while the nephroproct is situated more posteriorly (Fig. 2D). The cerata are entirely smooth, somewhat laterally compressed and are arranged in 5 to 6 closely packed longitudinal rows which are irregularly arranged. Within each ceras (Fig. 3A) are 2 to 4 irregular lobes of the digestive gland which branch near the middle of the ceras. The anteriormost cerata lack extensions of the digestive gland. The ground colour is translucent white. Opaque white markings may be present or absent on the rhinophoral lamellae, inter-rhinophoral crest and 4 ANNALS OF THE SOUTH AFRICAN MUSEUM along the dorsolateral margin of the foot. The cerata are tipped with an apical band of opaque white guanine crystals which may have a bluish tinge. The digestive gland within the cerata varies from chocolate brown to brick red or red-orange. Digestive system (Fig. 3B) Near the opening of the mouth, numerous small, simple oral glands are present on the surface of the outer lips of the buccal mass. The buccal mass is large and muscular with an ovoid opening. A pair of large, highly dendritic Fig. 3. Janolus capensis Bergh, 1907. A. Ceras dissected to show branching of digestive gland, scale 1,0 mm. B. Digestive system, scale 1,0 mm. SOUTH AFRICAN JANOLIDAE =) salivary glands extends anteriorly from the stomach and enters the buccal mass near its anterior limit via glandular ducts. The salivary glands are thick at their insertion, taper sharply and again expand into the dendritic portion. The oesophagus is wide throughout its length and enters the highly ridged stomach. Three major branches of the digestive gland emerge from the stomach. At the posterior limit of the stomach the intestine curves to the right and continues posteriorly, terminating at the medial anus. A well-developed anal gland surrounds the anus. The buccal mass is large and muscular with an oblong opening (Fig. 4A). The paired jaws (Fig 4B) are large and strong. The inner masticatory border (Fig. 5A) is thickened and entirely smooth. There is a raised, arched portion which strengthens the masticatory edge. The radula (Figs. 5B, 6) is broad and well developed. There may be from 17 to 21 rows of teeth with 26 to 42 lateral teeth on each side of the narrow, linear rachidian tooth. The hook-shaped lateral teeth (Fig. 6B) are smooth and sharply arched. Laterally from the centre of the radula, the lateral teeth increase in size until about one-third of the breadth of that half of the radula, at which point they again begin to diminish in size for the remaining outer two-thirds. Central nervous system All the major ganglia of the central nervous system (Fig. 7) are situated anteriorly in the circumoesophageal nerve ring. The cerebral and pleural ganglia are almost entirely fused, forming a pair of ganglionic masses separated by a short commissure. The anteriormost nerves on the dorsal surface of the Fig. 4. Janolus capensis Bergh, 1907. A. Buccal mass, scale 1,0 mm. B. Jaw, scale 1,0 mm. ANNALS OF THE SOUTH AFRICAN MUSEUM FAEes EAW , scale 300 wm between h of radula, scale 10 wm jaw between squares. icrograp is Bergh, 1907. A. Scanning electron tory border of lectron mi ica ing e micrograph of mast squares. B. Scann Fig. 5. Janolus capens SOUTH AFRICAN JANOLIDAE Fig. 6. Janolus capensis Bergh, 1907. A. Scanning electron micro- graph of central portion of radula. B. Scanning electron micrograph of lateral tooth from outer portion of radula. Scales 10 wm between squares. 8 ANNALS OF THE SOUTH AFRICAN MUSEUM Fig. 7. Janolus capensis Bergh, 1907. Central nervous system. Scale 0,5 mm cerebropleural ganglia bifurcate near the base. The outer branch innervates the rhinophore and a small rhinophoral ganglion is present near the apex of the nerve at the base of each rhinophore. The inner branch divides again with both branches innervating the inter-rhinophoral crest. The eyes are situated on short nerves which join the cerebropleural ganglia near their juncture with the somewhat smaller pedal ganglia. At the junction of each eye to the cerebro- pleural ganglia is a minute optic ganglion. Anterior to each buccal ganglion (Fig. 3B) is a minute gastro-oesophageal ganglion. The pedal ganglia are separated by a short commissure. Reproductive system (Fig. 8) The ovotestis consists of numerous lobes and gives rise to a slightly convoluted ampulla which subsequently narrows and bifurcates into the male wae | See Eee | Fig. 8. Janolus capensis Bergh, 1907. Reproductive system. Scale 4,0 mm. SOUTH AFRICAN JANOLIDAE 9 and female ducts. The vas deferens is highly convoluted and prostatic through- out its length. It terminates at an elongate, acutely pointed penis. The extremely elongate oviduct is muscular and gives rise to a very long receptacu- lum seminis and continues until it joins the albumen gland of the female gland mass at the female atrium. /n situ, the oviduct and receptaculum seminis are surrounded by the two lobes of the voluminous mucous gland. The membrane and albumen glands are significantly smaller than the mucous gland. Adjoining the oviduct and female gland mass, at the female atrium, is a small pyriform bursa copulatrix. Egg mass The egg mass (Fig. 9) is highly convoluted and corresponds to type B (Hurst 1967). There are 38 to 45 eggs per capsule. Natural history Janolus capensis is associated with, and feeds upon, several species of arborescent cheilostomatous ectoprocts, most commonly Menipea triseriata Busk, 1852, and Onchoporella buskii (Harmer, 1923). Mating individuals and egg masses are often found on ectoproct colonies as well. Janolus capensis is found commonly throughout the year along the Atlantic and Indian Ocean coasts of the Cape Peninsula from the intertidal to at least 40 m in depth. Janolus longidentatus sp. nov. Figs 1C, 10-17 Type material Holotype—SAM-A34883, 1 specimen, 3 m depth, Miller’s Point (34°14’S 18°29’E), 10 December 1980 Paratypes—SAM-A34884, 2 specimens, 10 m depth, Castle Rocks, False Bay (34°18’S 18°29’E), 1 October, 1980 SAM-—A34885, 1 specimen, 5 m depth, Miller’s Point, False Bay (34°14'S 18°29'E), 28 October 1980 Other material 3 specimens, intertidal, Clovelly, False Bay (34°05’S 18°26'E), 17 Decem- ber 1979 2 specimens, 10 m depth, Windmill Beach, False Bay (34°12'S 18°29’E), 11 July 1980 Etymology The epithet /ongidentatus refers to the elongate cusp of the lateral radular teeth which characterizes this species. 10 ANNALS OF THE SOUTH AFRICAN MUSEUM Fig. 9. Janolus capensis Bergh, 1907. A. Egg mass X 6. B. Single egg capsule x 50. SOUTH AFRICAN JANOLIDAE 11 External morphology Fully mature specimens (Fig. 1C) attain a maximum length of 20 mm. The slender body is widest anteriorly, tapering to an acute posterior margin of the foot. The rhinophores (Fig. 10A) are perfoliate with 9 to 12 complete or incomplete, transverse lamellae. The inter-rhinophoral crest (Fig. 10B) is highly convoluted, narrowest in the middle. The anus is situated mid-dorsally near the posterior end of the notum. The mouth is located centrally on the ventral side of the rounded head (Fig. 10C). A pair of short, blunt oral tentacles extend from either side of the head. The anteriorly rounded foot possesses a deep transverse groove at its anterior limit. The separate male and female gonopores are situated laterally in the anterior third of the right side of the body, while the nephroproct is located more posteriorly, but within the anterior half of the body (Fig. 10D). The entirely smooth cerata are long and slender, arranged in 4 or 5 irregular, longitudinal rows. Within each ceras there are 2 main branches of the digestive gland which usually branch again above the base of the ceras (Fig. 11). The anteriormost cerata lack extensions of the digestive gland within them. The general body colour of living animals is translucent white, with a pink or blue cast. Varying amounts of opaque white pigment are present on the rhinophores and at the apices of the cerata. The digestive gland within the cerata is chocolate brown to red-orange. Digestive system Near the mouth and surrounding the outer lips of the buccal mass is a narrow ring of small, simple oral glands. The buccal mass (Fig. 12A) is somewhat dorsoventrally compressed, muscular, with a circular opening. The paired salivary glands are very thin anteriorly and expand posteriorly into the dendritic, glandular portion with numerous ramifications. The oesophagus is a straight glandular tube which expands into the corrugated, saccate stomach. From the stomach arise three major branches of the digestive gland: two branches emanate from the anterodorsal portion of the stomach and give rise to the left and right anterior digestive branches; the third branch arises at the ventral portion of the stomach and branches to the right and left posterior digestive branches. The intestine emerges from the posterior portion of the stomach and curves to the right, continuing posteriorly to the anus. The anus is surrounded by a large anal gland. The jaws (Figs 12B, 13A) are of moderate thickness, angular and termin- ate posteriorly in a rather acute point. The masticatory border consists of 7 or 8 large, rounded denticles. The radula (Figs 13B, 14) is broad, consisting of 18 to 23 rows of teeth. There are 19 to 26 gradually arched, edenticulate lateral teeth on each side of the somewhat broad rachidian tooth. In the ten specimens examined, the rachidian teeth all possess 6 to 10 minute striations on each side of the base of the elongate central cusp. The lateral teeth have an elongate cusp 2 ANNALS OF THE SOUTH AFRICAN MUSEUM Fig. 10. Janolus longidentatus sp. nov. A. Rhinophore, scale 0,5 mm. B. Inter-rhinophoral crest, scale 0,5 mm. C. Ventral view of head, scale 1,5 mm. D. Lateral view, scale 2,0 mm. Figs 2A—B, D are drawn from preserved material, Fig. 2C is from living material. SOUTH AFRICAN JANOLIDAE 13 Fig. 11. Janolus longidentatus sp. nov. Cerata dissected to show branching of digestive gland. Scale 1,0 mm. Fig. 12. Janolus longidentatus sp. nov. A. Buccal mass, scale 1,0 mm. B. Jaw, scale 1,0 mm. 14 ANNALS OF THE SOUTH AFRICAN MUSEUM Fig. 13. Janolus longidentatus sp. nov. A. Scanning electron micrograph of masticatory border of jaw, scale 300 wm between squares. B. Scanning electron micrograph of radula, scale 10 wm between squares. SOUTH AFRICAN JANOLIDAE Bee sa ae | Fig. 14. Janolus longidentatus sp. nov. A. Scanning electron micrograph of central portion of radula, scale 30 wm. B. Scan- ning electron micrograph of lateral teeth from outer portion of radula, scale 30 wm between squares. 15 16 ANNALS OF THE SOUTH AFRICAN MUSEUM and increase in length towards the outer edges of the radula. Only the outermost teeth decrease in length. Central nervous system (Fig. 15) The major ganglia are situated in the circumoesophageal nerve ring. The cerebral and pleural ganglia are largely distinct. The anteriormost nerve of the dorsal surface of the cerebral ganglia bifurcates near its base with the inner branch innervating the inter-rhinophoral crest and the outer branch innervating the rhinophores. There is a distinct rhinophoral ganglion near the apex of the rhinophoral nerve. The eyes are situated at the apex of short optic nerves which join the central nervous system via small optic ganglia at the junction of the cerebral and pleural ganglia. Extending from the cerebral ganglia are the buccal ~ nerves which join the round buccal ganglia. Anterior to each buccal ganglion is a minute gastro-oesophageal ganglion. The pedal ganglia are long with an elongate commissure between them. Reproductive system (Fig. 16) The follicles of the ovotestis are numerous and arranged into distinct lobes. The thick ampulla has 2 or 3 convolutions and narrows to the bifurcation of the vas deferens and oviduct. The glandular prostatic vas deferens is short with few convolutions. It enlarges into a much thicker penis sac which is recurved to an acutely pointed unarmed penial papilla. The muscular oviduct is embedded Fig. 15. Janolus longidentatus sp. nov. Central nervous system. Scale 1,0 mm. SOUTH AFRICAN JANOLIDAE i7/ Fig. 16. Janolus longidentatus sp. nov. Reproductive system. Scale 1,0 mm. between two lobes of the mucous gland and branches near the middle of its length to a short, bulbous receptaculum seminis. The major branch continues distally to its connection with the albumen gland and female gonopore. The narrow, linear bursa copulatrix joins the oviduct near the junction of the female gland mass, at the female atrium. Egg mass (Fig. 17) The egg mass is a low, flat spiral consisting of 3 or 4 whorls and corresponds to type B (Hurst 1967). There are 5 to 9 eggs per capsule. Natural history Janolus longidentatus feeds upon the cheilostomatous ectoproct Menipea triseriata Busk. It is commonly found in False Bay but has not been encoun- tered on the Atlantic coast of the Cape Peninsula. J. longidentatus appears to have a seasonal distribution. Specimens have not been found from February to July and appear to be present largely in the winter and spring months. Specimens have been found from the intertidal to 30 m depth. Bonisa gen. nov. Diagnosis Body stout, ovoid in outline. Rhinophores perfoliate. Inter-rhinophoral crest low, triangular with few convolutions. Cerata smooth, caducous, not 18 ANNALS OF THE SOUTH AFRICAN MUSEUM Fig. 17. Janolus longidentatus sp. nov. A. Egg mass X 6. B. Single egg capsule x 100. SOUTH AFRICAN JANOLIDAE 19 containing branches of digestive gland. Gonopore near middle of body with nephroproct in posterior third of body. Oral glands small, simple. Salivary glands well developed with basal bulb. Digestive gland of three lobes surround- ing stomach, not extending into notum or cerata. Anal glands absent. Jaws thick with smooth masticatory border. Rachidian and lateral teeth weakly to strongly denticulate. Pedal ganglia lateral to well separated cerebral and pleural ganglia. Optic nerves elongate. Reproductive system hermaphroditic, andro- diaulic. Vas deferens short, prostatic, expanding into blunt, corrugated penis. Receptaculum seminis large, muscular and adjacent to gonopore. Bursa copula- trix small, linear and joining receptaculum seminis at female gonopore. Type species Bonisa nakaza sp. nov. Etymology Bonisa is named for my wife Bonnie Isabel Julien Gosliner. Bonisa nakaza sp. nov. Figs 1D-F, 18-25 Type material Holotype—SAM-A34886, 20 m depth, off Llandudno (34°01'S 18°20’E), ZaeDecember 1979. Paratypes—SAM-A34887, 4 whole specimens, 20 m depth, off Llandudno (34°01’S 18°20’E), 23 December 1979 Other material 1 specimen, 10 m depth, Oudekraal (33°59'S 18°21’E), 13 January 1980 2 specimens, 10 m depth, Castle Rocks, False Bay (34°18’S 18°29’E), 17 January 1980 Etymology Nakaza is a Zulu word meaning ‘to adorn with beautiful colours’, and the name is based on the bright coloration typical of this species. External morphology Fully mature specimens may reach a length of 100 mm. The body is broad, roughly oval in outline. The rhinophores (Fig. 18A) are perfoliate with 14 to 23 transverse lamellae. The slightly convoluted inter-rhinophoral crest (Fig. 18B) is triangular in shape, widest anteriorly. The mid-dorsal anus is located near the posterior end of the notum. The broad head (Fig. 18C) is round in shape, with a secondary, anterior lobe. The paired oral tentacles are moderately elongate and situated on either side of the head. The foot is transversely grooved at its 20 ANNALS OF THE SOUTH AFRICAN MUSEUM Fig. 18. Bonisa nakaza gen. et. sp. nov. A. Rhinophore, scale 1,0 mm. B. Inter-rhinophoral crest, scale 1,0 mm. C. Ventral view of head, scale 1,0 mm. D. Lateral view, scale 5,0 mm. Figs 2A—B, D are drawn from preserved material, Fig. 2C is from living material. SOUTH AFRICAN JANOLIDAE 71)\ anterior limit. The separate male and female gonopores are located near the middle of the right side of the body (Fig. 18D). The nephroproct is situated more posteriorly, in the posterior third of the animal. The elongate, readily caducous cerata are smooth and cylindrical. They are arranged in 16 to 21 diagonal rows per side with 8 to 10 cerata per row. Within the cerata a central muscular duct is present but this contains no extension of the digestive gland. At the base of each ceras is a ganglion which bifurcates into two nerves which extend distally into the ceras (Fig. 19A). The living animals are variably and brightly coloured. (Fig. 1D-F). The general body colour ranges from translucent white to yellow or yellow-orange. Yellow or orange pigment is particularly concentrated on the anterodorsal surface. The rhinophores possess varying amounts of yellow or orange pigment. The cerata are strikingly variable in their coloration. On the surface they are most commonly yellow, basally, with varying amounts and shades of blue pigment on the more distal portion. Some specimens possess a small subapical concentration of black or dark-blue pigment within the ceratal ducts. Other “ae " Bo FRAN \ y ay, 9 By." Ms aerate aa LA eri ORO POS OR ‘are NRE Soins d oon Ke 3a one SESH EA AOR BOIS AS \ 78 a rn Wes — iD wi 62 Ss (| LY PNR Na 4 4 r£5 = ~, Fig. 19. Bonisa nakaza gen. et sp. nov. A. Ceras, scale 1,0 mm. B. Digestive system, scale 2,0 mm. 22 ANNALS OF THE SOUTH AFRICAN MUSEUM specimens, which entirely lack blue or yellow pigment, possess only an apical band of orange on each ceras and rhinophore. A few juvenile specimens have cerata which are pinkish with blue pigment and opaque white apices. Since adult specimens of different colour patterns have been observed copulating, they are considered to be conspecific. Digestive system (Fig. 19B) There is a moderately large ring of oral glands surrounding the outer lips. The massive buccal mass is highly muscular and occupies the anterior third of the body cavity. The paired salivary glands are well developed, extending from the posterior limit of the oesophagus. Near their origin there is a large, spherical expansion of the salivary gland duct which probably functions as a vestibule for storage of secretory products. The oesophagus is short and expands into a thin-walled stomach. On the dorsal surface of the stomach is a large corrugated portion. Surrounding the stomach are three major lobes of the highly dendritic digestive gland. A digestive lobe joins the stomach on both the anterolateral sides of the stomach with a third lobe connecting to the antero- ventral portion. The intestine emerges from the posteroventral portion of the stomach and continues posteriorly to the anus. The anus is not surrounded by anal glands. The buccal mass (Fig. 20A) has an oblong opening. The jaws (Fig. 20B) are thick and broad with a smooth masticatory border (Fig. 21) which is supported by an elevated arch of chitin. The outer borders of the jaws are sharply indented. The radula (Fig. 22) contains 21 to 46 rows of teeth with 7 to Fig. 20. Bonisa nakaza gen. et sp. nov. A. Buccal mass, scale 1,0 mm. B. Jaw, scale 1,0 mm. SOUTH AFRICAN JANOLIDAE 23 wy 7 % ; es “ “ 4 a a i thy ie i pS : 4. & ass me ame see =e Fig. 21. Bonisa nakaza gen. et sp. nov. Scanning electron micrograph of masticatory border of jaw. Scale 100 wm between squares. 33 laterals on each side of the rachidian teeth. The rachidians possess 2 to 5 small or elongate denticles. The lateral teeth possess 2 to 7 denticles on each side of the elongate central cusp. Central nervous system (Fig. 23) The major ganglia of the central nervous system are situated in the circumoesophageal nerve ring. The cerebral ganglia are closely appressed to each other without a distinct commissure and are joined posteriorly by the distinct pleural ganglia. The anteriormost dorsal nerve of the cerebral ganglion innervates the rhinophores and divides near the middle of its length to give rise to a very thin branch which innervates the inter-rhinophoral crest. The eyes are situated at the distal end of the elongate optic nerve which joins the junction of the cerebral and pleural ganglia via a short optic ganglion. Extending anteriorly from the cerebral ganglia are the paired buccal nerves which enlarge into the buccal ganglia on the surface of the buccal mass. Ventral to the oesophagus, the buccal ganglia are joined by a commissure of moderate length and anter- iorly each gives rise to a smail gastro-intestinal ganglion. The pedal ganglia are situated laterally to the cerebral and pleural ganglia rather than posteriorly, and are connected by an elongate commissure. 24 ANNALS OF THE SOUTH AFRICAN MUSEUM Fig. 22. Bonisa nakaza gen. et sp. nov. A. Scanning electron micrograph of central portion of radula. B. Scanning electron micrograph of lateral teeth from outer portion of radula. Scales 30 um between squares. SOUTH AFRICAN JANOLIDAE 25 =, Fig. 23. Bonisa nakaza gen. et sp. nov. Central nervous system. Scale 0,5 mm. Reproductive system (Fig. 24) The ovotestis consists of numerous follicles which are united into an ovoid mass. The preampullary duct emerges from the anterior end of the ovotestis and expands into a highly convoluted, thickened ampulla. The ampulla narrows anteriorly and divides into the oviduct and vas deferens. The vas deferens is short, thick, and muscular, and expands into the massive penis sac. The penial papilla is club-shaped and strongly corrugated. The oviduct is muscular and joins the large receptaculum seminis near its base. Traversing the base of the receptaculum seminis is a small, linear bursa copulatrix which joins the recepta- culum near the female gonopore. At the base of the receptaculum sac is a thick duct which connects it with the small albumen and membrane portions of the female gland mass. These glands join the massive mucous gland, which forms the bulk of the female gland mass. Egg mass (Fig. 25) The egg mass is about 20 to 30 mm high, consists of numerous convolu- tions and corresponds to type B (Hurst 1967). There are one or two eggs per capsule. The egg mass is frequently deposited on the finger-like projections of gorgonians. Natural history Bonisa nakaza is found in the shallow subtidal, at depths of 3 to 30 m on both the Atlantic and Indian Ocean coasts of the Cape Peninsula. It is 26 ANNALS OF THE SOUTH AFRICAN MUSEUM Fig. 24. Bonisa nakaza gen. et sp. nov. Reproductive system. Scale 4,0 mm. associated with and feeds upon the heavily calcified arborescent cheilostoma- tous ectoproct, Tubucellaria levinseni Canu & Bassler, 1930. DISCUSSION OF THE JANOLIDAE Within the suborder Arminacea, the tribe Pachygnatha is subdivided into three families, the Madrellidae, Dironidae, and Janolidae (Franc 1968). The Janolidae have been separated from the other two families by having a broad rather than narrow radula, a medial rather than lateral anus, and by having extensions of the digestive gland into the cerata. With the discovery of Bonisa nakaza in this study, the diagnosis of the Janolidae must be expanded to include forms with or Without extensions of the digestive gland into the cerata. The family Janolidae has been previously subdivided into five genera. Proctonotus Alder & Hancock, 1844, and Caldukia Burn & Miller, 1969, lack an inter-rhinophoral crest. Caldukia can be separated from Proctonotus by its distinctively stout rhinophores with transverse lamellae and its significantly narrower radular ribbon. Of the genera which possess an inter-rhinophoral crest Galeojanolus Miller, 1971, can be separated from Janolus Bergh, 1884, and Antiopella Hoyle, 1902, by the helmet-like appendage extending anteriorly from the head. Bonisa gen. nov. must be added to this group and can be readily separated from the other janolids by the fact that the digestive gland surrounds the stomach rather than entering the notum and cerata. In several respects Bonisa nakaza most closely approaches Galeojanolus ionnae Miller, 1971. Both species have an ovoid body shape, similar coloration, perfoliate rhinophores, a low, broad inter-rhinophoral crest with few convolutions, jaws with a smooth masticatory border, a moderately broad radula with strongly denticulate teeth and a diaulic reproductive system. Both species lack anal glands. However, there are several significant differences which distinguish the species. Galeo- SOUTH AFRICAN JANOLIDAE Fig. 25. Bonisa nakaza gen. et sp. nov. A. Egg mass X 3. B. Several egg capsules x 100. Di 28 ANNALS OF THE SOUTH AFRICAN MUSEUM janolus ionnae possesses an elongate helmet-like appendage on the anterior end of the head which may be used in capturing mobile prey (Miller 1971), while there is a short lobe anterior to the head of Bonisa nakaza. G. ionnae has papillate, inflated cerata, while they are smooth and cylindrical in B. nakaza. B. nakaza lacks complex oral glands and is devoid of digestive gland ducts in the cerata which are both present in G. ionnae. The reproductive s¥stem of G. ionnae, while not fully mature, has a conical, tapered penis and an elongate vas deferens, while B. nakaza has a short vas deferens and a blunt, corrugated penis similar in shape to that described for Janolus cristatus (Schmekel 1970). Miller described a bursa copulatrix in G. ionnae, while B. nakaza possesses both a large receptaculum seminis and a small bursa copulatrix situated near the gonopore. These differences, particularly those of the digestive system, ' warrant generic as well as specific separation. It is difficult to separate Antiopella and Janolus. Several authors have united the genera (Thiele 1931; Franc 1968; Thompson & Brown 1976), while others have maintained them as distinct (Eliot 1906; Pruvot-Fol 1954; Marcus 1958; Burn & Miller 1969; Baba & Abe 1970; Miller 1971). Pruvot-Fol (1954) stated that Antiopella may be separated from Janolus by virtue of the fact that the component species possess a denticulate instead of smooth masticatory border of the jaw, smooth instead of papillate cerata, and branched instead of undivided ducts of the digestive gland within the cerata. However, several species of janolids (Table 1) possess intermediate or variable character states or individual characters. They may also possess a mixture of ‘Antiopella’ and ‘Janolus’ characters. Marcus (1958: 40) noted that the structure of the cerata in the type species of Janolus, J. australis, is questionable. Several workers who have suggested the separation of Antiopella from Janolus have stated that jaw denticulation alone should not warrant generic separation (Pruvot-Fol 1954; Marcus 1958). The fact that Janolus capensis and J. longidentatus are very similar in many aspects of their external and internal anatomy (present study) but differ in jaw denticulation supports the view that this character alone should not be utilized for generic separation in the Janolidae. The division of the genital ducts into a diaulic or triaulic configuration varies significantly within Janolus and Antiopella. The reproductive anatomy is described for eight of the fifteen species described in both genera. In three species, A. mucloc (Marcus 1958), Janolus comis (Marcus 1958), and J. toyamensis (present TABLE | Antiopella and Janolus characteristics in three species of janolids. Species Ceratal epithelium Ceratal ducts Jaws Janolus toyamensis smooth or with branched (A) large few tubercles (A or I) denticles (1) Janolus capensis smooth (A) branched (A) smooth (J) Janolus fuscus smooth (A) branched (J) denticulate (A) A—Antiopella I—Intermediate J—Janolus SOUTH AFRICAN JANOLIDAE 29 study), the reproductive system is triaulic. In A. cristata (Schmekel 1970), J. hyalinus (Schmekel 1970), J. fuscus (present study), J. capensis (present study), and J. longidentatus (present study) a diaulic configuration is present. The reproductive system of the type species of Janolus, J. australis, is incompletely described. There is no correlation between division of the genital ducts and jaw denticulation. There is, therefore, little basis for separating Antiopella Hoyle, 1902, from Janolus Bergh, 1884, and Antiopella is here regarded as a junior subjective synonym of Janolus on the basis of priority. KEY TO THE GENERA OF THE JANOLIDAE ime izestive sland surrounding Stomach . 22... . i... ite tnt eee ee eee eee Bonisa Digcstivercland extending into notum and cerata ’. 97 .-05. 03445: oes e sas saves anne 2 PRC -oMInOphoral Crest PLESENE. 25 ow Mee ee ik we Tae eye ee ahaa gemebes 3 Inte roninophnoralccrest:abSemt 2s. Ps e256 eet tee oe oa ele al wes oe hes eda ew dees 4 Sc acawiininelmet-lke EXtenSiOM 2. oan eos Meee ne sek be cee meme eats Galeojanolus Hela NWAGMOUIREKCENGIOM. 5 oes). nj sesepane ao Heel eyerg ales Bt Ht ge ea bs eo nee Janolus 4 Rhinophoral club stout with transverse lamellae, radula6.1.6................ Caldukia Rhinophores with inconspicuous lamellae, radula broad.................. Proctonotus The following species are considered to constitute the family Janolidae (“indicates type species): Bonisa gen. nov. 1. *Bonisa nakaza sp. nov. Caldukia Burn & Miller, 1969 2. *Caldukia affinis (Burn, 1958) Proctonotus? affinis Burn, 1958: 32, fig. 8, pl. 7 (fig. 15). Caldukia affinis (Burn, 1958), Burn and Miller, 1969: 23, figs 1-2, pl. 2. 3. Caldukia albolineata Miller, 1970 Caldukia albolineata Miller, 1970: 279, figs 1-10. 4. Caldukia rubiginosa Miller, 1970 Caldukia rubiginosa Miller, 1970: figs 11-21. Galeojanolus Miller, 1971 5. *Galeojanolus ionnae Miller, 1971 Galeojanolus ionnae Miller, 1971: 491, figs 1-4. Janolus Bergh, 1884 6. *Janolus australis Bergh, 1884 Janolus australis Bergh, 1884: 19, pl. 8 (figs 15-22), pl. 9 (figs 6-8). 30 ANNALS OF THE SOUTH AFRICAN MUSEUM 7. Janolus barbarensis (Cooper, 1863) Aeolis barbarensis Cooper, 1863: 59. Janolus coeruleopictus Cockerell & Eliot, 1905: 48, pl. 8 (figs 12-16). O’Donoghue, 1922: 141. Janolus barbarensis (Cooper, 1863), O’Donoghue, 1922: 141. Antiopella aureocincta MacFarland, 1966: 303, pl. 57 (figs 1-5), pl. 63 (figs 13-30), pl. 64 (figs 117) Roller 197053725 in) pant- 8. Janolus capensis Bergh, 1907 Janolus capensis Bergh, 1907: 90, pl. 7 (figs 6-21). 9. Janolus comis Marcus, 1955 Janolus comis Marcus, 1955: 170, pl. 25 (figs 226-236). 10. Janolus cristatus (Chiaje, 1841) Eolis cristatus Chiaje, 1841: pl. 88 (figs 1-12). Janus spinolae Verany, 1845: 24, pl. 2 (fig. 9). Pruvot-Fol, 1954: 375. Antiopa splendida Alder & Hancock, 1848: 190. Alder & Hancock, 1851. Antiopella cristata (Chiaje, 1841), Hoyle, 1902: 214. Janolus cristatus (Chiaje, 1841), O’Donoghue, 1924: 1. 11. Janolus flagellatus Eliot, 1906 Janolus flagellatus Eliot, 1906: 374. (Likely a junior synonym of J. hyalinus (Alder & Hancock, 1854). Eliot (1910) considered this as an uncertain species. It was distinguished from J. hyalinus by possession of a flagellar penis which has subsequently been described in J. hyalinus (Schmekel 1970).) 12. Janolus fuscus O’Donoghue, 1924 Janolus fuscus O’Donoghue, 1924: 16, pl. 2 (figs 18-20). Antiopella aureocincta Johnson & Snook, 1927: 500. Nomen nudum, Steinberg, 1963: 66. Antiopella aureotincta MacGinitie & MacGinitie, 1949: 135. Nomen nudum, Steinberg, 1963: 66. Antiopella aureocincta MacFarland, 1966: 303, pl. 57 (figs 1-5), pl. 63 (figs 13-30), pl. 64 (figs 11-17). Roller, 1970: 372, in part. 13. Janolus hyalinus (Alder & Hancock, 1854) Antiopa hyalina Alder & Hancock, 1854a: 105. Janolus hyalinus (Alder & Hancock, 1854), Eliot, 1906: 374. 14. Janolus indicus (Eliot, 1909) Antiopella indica Eliot, 1909: 143. Janolus indicus (Eliot, 1909), comb. nov. 15. Janolus longidentatus sp. nov. 16. Janolus mirabilis Baba & Abe, 1970 Janolus mirabilis Baba & Abe, 1970: 65, figs 2-3. 17. Janolus mucloc (Marcus, 1958) Antiopella mucloc Marcus, 1958: 37, figs 62-71. Janolus mucloc (Marcus, 1958), comb. nov. 18. Janolus novozealandicus (Eliot, 1907) Antiopella novozealandica Eliot, 1907: 331. Janolus novozealandicus (Eliot, 1907), comb. nov. SOUTH AFRICAN JANOLIDAE Sil 19. Janolus praeclarus (Bouchet, 1975) Antiopella praeclara Bouchet, 1975: 127, pl. 1 (fig. 3), fig. 5. Janolus praeclarus (Bouchet, 1975), comb. nov. 20. Janolus toyamensis Baba & Abe, 1970 Janolus toyamensis Baba & Abe, 1970: 63, fig. 1. Proctonotus Alder & Hancock, 1844 21. *Proctonotus mucroniferus (Alder & Hancock, 1844) Venilia mucronifera Alder & Hancock, 1844: 163, pl. 2. Proctonotus mucroniferus (Alder & Hancock, 1844) Alder, 1844: 407. Zephyrina pilosa Quatrefages, 1844: 130, pl. 3 (fig. 1), pl. 4 (fig. 1), pl. 5 (figs 1-2), pl. 6 (figs 1, 11-12). Pruvot-Fol, 1954: 372. Morphological variability Janolus varies in its external and internal morphology. The major morpho- logical features and their character states in the fifteen described species of Janolus are summarized in Table 2. Many taxa are incompletely described, including the type species, but the available morphological data does permit a review of the variability within the genus. All described species have a simply rounded head, perfoliate or, exceptionally, papillate (J. comis Marcus, 1955) rhinophores, an inter-rhinophoral crest, and cerata which contain ducts of the digestive gland. The inter-rhinophoral crest may be small and rectangular or may be highly elaborate as in J. cristatus (Alder & Hancock 18546), J. capensis (present study), J. novozealandicus (Eliot 1907), and J. longidentatus (present study). The cerata are papillate in J. comis (Marcus 1955), J. hyalinus (Alder & Hancock 1854b), and J. mirabilis (Baba & Abe 1970), and smooth in the remainder of the species. The cerata of J. toyamensis are variable, with scattered, small papillae (Baba & Abe 1970) or with an entirely smooth surface (present study). The digestive gland may extend only within the basal half of the cerata in Janolus comis (Marcus 1958), J. hyalinus (Bergh 1904), and Galeojanolus ionnae (Miller 1971), or extend to the apex of the cerata in the remaining species of Janolus. Within the cerata the digestive gland may be branched or unbranched. In species with a branched digestive gland, the splitting may occur apically in J. cristatus (Alder & Hancock 185456) and J. barbarensis (MacFarland 1966), near the middle in J. mucloc (Marcus 1958), J. praeclarus (Bouchet 1975), J. toyamensis (Baba & Abe 1970), and J. capensis (present study), or basally in J. longidentatus (present study). In most species of Janolus all cerata contain extensions of digestive gland except J. hyalinus (Eliot 1910), J. comis (Marcus 1958), J. capensis (present study), and J. longidentatus (present study) where the anteriormost cerata are devoid of digestive gland tissue. It appears that in all species of Janolus that have been studied, simple oral glands and elaborate salivary glands are present and the digestive gland occurs as three major branches from the stomach. Anal glands appear to be present in all species except J. comis (Marcus 1955) and J. fuscus (present study). a2 ANNALS OF THE SOUTH AFRICAN MUSEUM TABLE 2 Comparative morphology of Janolus. Species of Ceratal Ceratal 5 Janolus Colour epithelium ducts Jaws Radular teeth J. australis unknown smooth ? unknown smooth rachidian and = laterals smooth J. barbarensis . translucent white; cerata smooth branched unknown rachidian smooth, w with gold and blue laterals: inner 2 denticulate,| | outer smooth J. comis light brown with dark- papillate unbranched smooth rachidian denticulate, 4 brown stipples laterals: denticulate, variable} * J. cristatus cream or brown with smooth branched at 8-12 denticles rachidian and laterals opaque white cerata apex smooth, rarely denticulate J. flagellatus unknown unknown unknown smooth rachidian and laterals smooth J. fuscus translucent white with smooth unbranched 8-14 denticles rachidian denticulate, : brown lines, cerata with laterals: inner 2 denticulate,| | yellow and white outer smooth J. hyalinus cream with red-brown papillate unbranched smooth rachidian and laterals blotches; cerata dotted with smooth or finely denticulate white and brown J. indicus translucent with red-brown unknown unknown denticulate rachidian and laterals spots, cerata grey-green smooth J. mirabilis yellow with brown and papillate unbranched? smooth rachidian, denticulate, Opaque white; cerata white laterals: inner 3 denticulate, at apex outer 2 smooth J. mucloc . transparent white with smooth branched, 8 denticles rachidian denticulate, | opaque white lines; cerata bifurcate laterals: inner denticulate, with orange spot and below outers smooth opaque white lines middle J. novozealandicus grey with purple stripe and smooth unbranched indistin-t rachidian and laterals spots; cerata with purple denticulation smooth spots J. praeclarus orange with opaque white; smooth branched 6 denticles rachidian and laterals cerata with yellow bands smooth and bluish-white apex J.toyamensis . yellow with white and few minute branched 2 or 3 large rachidian and laterals brown markings; cerata papillae denticles smooth with yellow spot and red- brown apex J. capensis translucent white with smooth branched smooth rachidian and laterals Opaque white; cerata red- smooth brown with opaque white apex J. longidentatus translucent white with smooth branched 7-8 denticles rachidian striate, ; Opaque white; cerata red- basally laterals smooth brown with opaque white apex { | | 21 x 21.1.21 SOUTH AFRICAN JANOLIDAE 33 Receptaculum Bursa Prostatic Reproductive Radular formula seminis copulatrix Penis vas deferens ducts References 24 x 30-54.1.30-54 unknown unknown thicker unknown unknown Bergh 1884 than vas deferens 6X27 127 unknown unknown unknown unknown unknown Cockerell & Eliot 1905 MacFarland 1966, in part 16 x 12—20.1.12—20 proximal, absent conical, proximalto triaulic Marcus 1955 25 x 41.1.41 serial flagellar penis only Marcus 1958 30 x 40.1.40 proximal, minute, blunt throughout diaulic Alder & Hancock 1851 24 x 33.1.33 semi-serial distal Alder & Hancock 1855 Schmekel 1970 Thompson & Brown 1976 15 x 20.1.20 unknown large, unknown unknown unknown Eliot 1906 elliptical PAS C22 N22 proximal, distal conical throughout diaulic O’Donoghue 1924 ZO25. 1625 semi-serial MacFarland 1966, in part present study 15 x 11—13.1.11-13 small, absent elongate, throughout diaulic Alder & Hancock 1854b proximal, flagellar Alder & Hancock 1855 Bergh 1888, Eliot 1906, semi-serial Schmekel 1970 Pip sletesil unknown unknown unknown unknown unknown Eliot 1909 DO x5 15 unknown unknown unknown unknown unknown Baba & Abe 1970 18 x 24.1.24 proximal, absent thick throughout triaulic Marcus 1958 semi-serial within female gland mass PX Silos unknown unknown short unknown unknown Eliot 1907 20 x 28.1.28 unknown unknown unknown unknown unknown Bouchet 1975 20 x 10-—20.1.10-20 proximal absent elongate, throughout triaulic Baba & Abe 1970 OX 25S Ae25 flagellar present study 20 x 42.1.42 elongate, distal, elongate, throughout diaulic Bergh 1907 18 x 41.1.41 proximal, minute flagellar Barnard 1927 17 x 26.1.26 semi-serial present study LETS< AAT a short, distal, conical short diaulic present study 22) 26:1226 proximal, minute 20 x 20.1.20 semi-serial 32 ANNAL S OF THE SOUTH AFRICAN MUSEUM SOUTH AFRICAN JANOLIDAE TABLE 2 Comparative morphology of Janolus. Ceratal Species of pit ducts Jaws Radular teeth Receptaculum Bursa Prostatic Reproductive Janolus Colour ; aan Radular formula seminis copulatrix Penis vas deferens ducts References . th ? unknown smoot rachidian and : J. australis unknown estel) laterals smooth 24 x 30-54.1.30-54 = unknown unknown pueneD unknown unknown Bergh 1884 an vas deferens as anched unknown rachidian smooth J. barbarensis . EransLiceus site ETA smooth brane! NaPeTAIee inner 2 denticul, 16 x 27.1.27 unknown unknown unknown unknown unknown Cockerell & Eliot 1905 with gold an outer smooth y MacFarland 1966, in part 5 5 i th achidi i z J. comis light brown with dark- papillate unbranched smoo HSE PANE, 16 x 12-20.1.12-20 —_— proximal, absent conical, Proximalto triaulic Marcus 1955 brown stipples » Variable 25x 41.1.41 serial flagellar penis only Marcus 1958 = i th branched at 8-12 denticles rachidian and laterals A A = J. cristatus cream or brown with sn100, A 30 x 40.1.40 proximal, minute, blunt throughout diaulic Al ancock 185 opaque white cerata pees smooth: rately eaten 24 x 33.1.33 semi-serial distal : ; Ree ee enee Schmekel 1970 = Thompson & Brown 1976 J. flagellatus . unknown unknown unknown smooth Aechistan and laterals 15 x 20.1.20 unknown Testes unknown unknown unknown Eliot 1906 elliptica’ J. fuscus translucent white with SH Rote ay CCRC eS ag Facclan denteua 21 x 2.1.22 proximal, distal conical throughout diaulic O'Donoghue 1924 brown lines, cerata with Bi nae 26 x 25.1.25 semi-serial MacFarland 1966, in part yellow and white S present study ; F ; i branched smooth rachidian and laterals aoa J. hyalinus cream with red-brown __ papillate wn S 15% 11-13.1.11-13 small, absent elongate, throughout diaul Al ancock 185: 6 blotches; cerata dotted with smooth or finely denticulate proximal, flagellar g) iaulic ALE - Pancake Las white and brown : f Bergh 1888, Eliot 1906, semi-serial Schmekel 1970 J. indicus . translucent with red-brown unknown unknown denticulate peehidian and laterals 21 31.1.31 unknown unknown unknown unknown unknown Eliot 1909 spots, cerata grey-green smoot . mirabilis ellow with brown and papillate unbranched? smooth rachidian, denticulate, ant Ie k k J. mirabilis seaehe De teoraial White laterals: inner 3 denticulat, 20 x 5.1.5 unknown unknown unknown unknown unknown Baba & Abe 1970 at apex outer 2 smooth J. mucloc . transparent white with smooth branched, 8 denticles rachidian denticulate, 18 x 24.1.24 proximal, absent thick throughout _ triaulic Marcus 1958 opaque white lines; cerata bifurcate laterals: inner denticulate, semi-serial with orange spot and below outers smooth within female opaque white lines middle gland mass J. novozealandicus grey with purple stripe and smooth unbranched indistin ot. rachidian and laterals 23 %37.1.37 unknown unknown short unknown unknown Eliot 1907 spots; cerata with purple denticulation smooth spots J. praeclarus . orange with opaque white; smooth branched 6 denticles rachidian and laterals 20 x 28.1.28 unknown unknown unknown unknown unknown Bouchet 1975 cerata with yellow bands smooth and bluish-white apex J. toyamensis . yellow with white and few minute branched 2 or 3 large rachidian and laterals 20 x 10-20.1.10-20 proximal absent elongate, throughout triaulic Baba & Abe 1970 brown markings; cerata papillae denticles smooth 19 x 25.1.25 flagellar present study with yellow spot and red- brown apex J. capensis translucent white with smooth branched smooth rachidian and laterals 20 x 42.1.42 elongate, distal, elongate, throughout diaulic Bergh 1907 opaque white; cerata red- smooth 18x 41.1.41 proximal, minute flagellar Barnard 1927 brown with opaque white 17 x 26.1.26 semi-serial present study apex J. longidentatus translucent white with smooth branched 7-8 denticles rachidian striate, 18 x 21.1.21 short, distal, conical short diaulic present study opaque white; cerata red- basally laterals smooth 22 x 26.1.26 proximal, minute brown with opaque white 20 x 20.1.20 semi-serial apex 21x 21.1.21 34 ANNALS OF THE SOUTH AFRICAN MUSEUM As mentioned in the discussion of the synonomy of Antiopella with Janolus the elaboration of the masticatory border and shape of the jaw are interspeci- fically variable (Fig. 26). Previous workers (Pruvot-Fol 1954; Marcus 1958) have emphasized the distinctness of a smooth versus denticulate masticatory border. However, Janolus toyamensis (Fig. 26C) possesses three large tubercles on the masticatory border and J. novozealandicus was described as having indistinct denticles (Eliot 1907). The radular teeth are variable within Janolus (Fig. 27) and some intraspe- cific variations exists. The rachidian teeth are denticulate in J. fuscus (MacFar- land 1966; present study), J. comis (Marcus 1958), J. mirabilis (Baba & Abe 1970), and J. mucloc (Marcus 1958), striate in J. longidentatus (present study), and smooth in the remaining species. The rachidian tooth has a broad base in species with denticulations or striations and is linear in species with smooth teeth. In J. fuscus (MacFarland 1966; present study), J. mirabilis (Baba & Abe 1970), and J. mucloc (Marcus 1958) the inner 1 to 3 lateral teeth are denticulate while the outer teeth are entirely smooth. J. cristatus generally has smooth lateral teeth (Alder & Hancock 1855), but exceptionally (Bergh 1874) they are denticulate. The presence or absence of denticles varies with age in J. comis (Marcus 1958) and J. hyalinus (Eliot 1906). Bouchet (1975) noted that the lateral teeth of J. praeclarus increase in size until the eighteenth row and then begin to diminish. In J. australis the innermost are largest (Bergh 1884). In J. cristatus (present study) and J. capensis (present study) the teeth increase in length on either side of the radula until about one-third of the breadth, and subsequently diminish. In J. capensis the radular morphology of fifteen speci- mens of various sizes did not vary significantly. In J. hyalinus (Eliot 1906) and J. novozealandicus (Eliot 1907), and in at least ten specimens of J. longidenta- tus (present study) the largest lateral teeth are found at or adjacent to the outer edge of the radula. The relative length of the base to the cusp of the lateral teeth varies considerably between species and appears to be useful in the separation of species. The central nervous system of Janolus cristatus was described (Alder & Hancock 1851) as containing distinct cerebral and pleural ganglia. Bergh noted that these ganglia are fused in Janolus australis (Bergh 1884) and J. capensis (Bergh 1907). The present study confirms this arrangement in J. capensis, while J. longidentatus possesses distinct ganglia as in J. cristatus. The morphology of the reproductive system has been fully described in five species of Janolus. In J. comis and J. mucloc (Marcus 1958) the arrangement is triaulic with a short uterine duct connecting the vagina and female gland mass near the separation of the oviduct and vas deferens from the ampulla. In J. cristatus and J. hyalinus (Schmekel 1970) the vagina joins the female gland mass only at the female atrium, producing an androdiaulic configuration. The present study has shown that J. toyamensis (Fig. 28G) has a triaulic arran- gement with a proximal receptaculum seminis but with no bursa copulatrix. J. capensis and J. longidentatus have a diaulic arrangement with a proximal SOUTH AFRICAN JANOLIDAE a5 Fig. 26. Variation in jaw morphology. A. Janolus fuscus O’Donoghue, 1924. B. Bonisa nakaza gen. et sp. nov. C. Janolus toyamensis Baba & Abe, 1970. D. Janolus longidentatus sp. nov. E. Janolus capensis Bergh, 1907. Scale 1,0 mm. 36 ANNALS OF THE SOUTH AFRICAN MUSEUM ss Fig. 27. Variation in the rachidian, first, seventh and twentieth lateral teeth in the Janolidae. A. Janolus capensis Bergh, 1907. B. Janolus longidentatus sp. nov. C. Janolus cristatus (Chiaje, 1841). D. Janolus fuscus O’Donoghue, 1924. E. Janolus toyamensis Baba & Abe, 1970. F. Bonisa nakaza gen. et sp. nov. Not drawn to scale. SOUTH AFRICAN JANOLIDAE 37 receptaculum seminis and a minute, distal bursa copulatrix. This configuration is also present in J. fuscus (Fig. 28F) contrary to that described by MacFarland (1966, pl. 64 (fig. 11)). The shape of the penis has been utilized as a characteristic for specific separation. Janolus flagellatus (Eliot 1906) was described as distinct from J. hyalinus because its elongate, flagellar penial papilla was thought to be differ- ent from the blunt papilla depicted by Bergh (1888) for J. hyalinus. Subsequent work by Schmekel (1970) has shown that the penis of J. hyalinus is flagellar when fully extended. The shape of the penis varies considerably within species depending on degree of extension. In comparing penial structure between species, it is imperative that descriptions be made from specimens with a fully extended penis. The egg mass of Janolus cristatus (Alder & Hancock, 1855: pl. 44 (figs 6—7)) is a short, convoluted ribbon of about one and a half whorls, with a single egg per capsule. A similar egg mass is apparently present in Janolus fuscus (O’Donoghue 1924). The egg mass of Janolus capensis (present study) is globose, consisting of many whorls and with 38 to 45 eggs per capsule, while that of J. Jongidentatus is a flat spiral consisting of 5—9 eggs per capsule. Janolus longidentatus differs from all other species of Janolus which possess denticles on the masticatory edge. J. mucloc (Marcus 1958) is triaulic and therefore differs from the diaulic J. Jongidentatus. J. praeclarus (Bouchet 1975), while much of its anatomy remains unknown, differs from J. longidenta- tus in its Orange versus translucent white coloration, smaller inter-rhinophoral crest, branching of the ceratal ducts near the middle rather than basally, more sharply arched radular teeth that are longest near the middle rather than at the outer edge of the radula, and jaws with 6 denticles rather than 7 or 8. Janolus longidentatus differs from J. fuscus in its coloration, more elaborate inter- rhinophoral crest, smooth inner lateral teeth, and presence of anal glands. Janolus longidentatus, although similar to J. cristatus in coloration, external and reproductive anatomy (Schmekel 1970), differs in several significant aspects. In J. longidentatus the ceratal ducts branch basally while they branch apically in J. cristatus. The rachidian teeth of J. Jongidentatus are broad with striations versus the smoothly linear form of J. cristatus. In J. cristatus the radular teeth are largest at the inner third of the radula, while they are broadest near the outer margin in J. longidentatus. The lateral teeth are broader and heavier in J. cristatus. The oviduct is short in J. cristatus and elongate in J. longidentatus. J. cristatus has a blunt club-shaped penis, while in J. /ongidentatus it is conical. Despite the difference in the elaboration of the masticatory border of the jaw, J. longidentatus most closely approaches J. capensis in its morphology. For this reason a morphological comparison of the three sympatric South African janolids is provided in Table 3. The South African species of janolids occur together in the shallow waters of False Bay. All three species can be found within a few centimetres of each other. Janolus capensis and J. longidentatus have been observed to feed upon ANNALS OF THE SOUTH AFRICAN MUSEUM 38 -6-S Sp-8E Zi 10 if ajnsdes/ss30 yisugy] jeursea Jo Joy1enb y70ys BUISEA SB SUC] Sev ‘d}e3U0[S a7e90es ‘aSIe] SIUIWIOS WNTNIe}d9901 SUOTINJOAUOD MOI yum poinjoauos ATYysTY yys1es “y10Ys SUdIOJOp SBA yeotuos Ieyjosey povesnis09 kat stuod jouN)sIp posnj youNSsIp eljsues jeinojd pue [eiqo1eo qjoours ‘dsno a3e3u0]9 YIM yoous ‘podeys-yooy aje[nouep ‘podeys-yooy Y199} [e19}e] e[NpeI JO UIsIVW 19j}NO IevoU B[Npel JO YIpIM jo ply) 4199} JSOWIOUUI sje1o}e] 1SoSIe] SUONPII}S YIIM ‘peoiq reoul] ‘YJOOUS sopPluep o31e] YIM ‘peolq YJ00} UvIpPIyoes Sopoluep g-/ yJOous YJOOUIS Japiog A1ojeonseul UONIISUI Je Ie[Npueys uOonJesuI je Jode}y IIOAIOSOI [eseq YIM spurs Arearyes punol ro 0(0) (@[0) Zuojqo sulusdo ynoul Apoqg Jo d]pprw Iesu Apoq jo o][pprul 0} Jo1193sod Apoq JO pity} 1o11a}sod yooido1ydou Apoq Jo pity} Jo1ajue Apog Jo 3[pprw edu Apog jo o]pprw re9u a1odouos aseq Ieau ‘iejnse1 ‘pylupenb J[pplu Ieou ‘Ie[NsoIM “pynnuw juasqe sjonp [eye100 suOINJOAUOS Aue ‘YsTY suolNjoAuos Aueu ‘Yysty SUONNJOAUOD MOJ ‘MOT soo Je1OYdouTYI-19}UI Jopusys no}s yno}js Apoq Snjvuap1suo] snjouve sisuadvo snjouve DZDYDU DSIUOg ‘sprjouel uevoLyy ynos jo Asojoydiow saneiedwod ¢ ATaV 39 SOUTH AFRICAN JANOLIDAE “Q[2OS 0} UMBIP JON ‘(Apnys yussoid) ‘Aou ‘ds j9 ‘uod DzDynU DSiuog “HY ‘(Apnys yuosoid) OL6I “9QV w eqeg sisuaupdo] snjouve ‘DH “(Apnys jussoid) p76, ‘onysou0d,.C snosn{ snjouvr ‘4 “(OL6T [OYOWYOS Joye) (TpET ‘oferD) snywjsi9 snjouvr “ZA “(OL6I JPYouyos Joye) (post “YoooueH 2%» Jppy) snuyody snjouve *q “(OLGT [OYewWYIS Joe) Ccoy ‘snoIe] suo snjouve *D *(Apnjs juasoid) -aou ‘ds snipjuapisuo] snjouvr ‘gq ‘(Apnys juasoid) (06, ‘YsiI9g sisuadvo snjouvs ‘YW ‘snjouvs ut Wa\sks dAONpoOidal Jo UONeLIeA °87 “SI d SA SA Pa Su ule 40 ANNALS OF THE SOUTH AFRICAN MUSEUM the same species of ectoproct, while Bonisa nakaza feeds on a more heavily calcified species. Fdod abundance does not appear to be a limiting factor as many ectoproct colonies are devoid of janolids. Observation of feeding behav- iour in Janolus capensis and J. longidentatus has produced no apparent differ- ences in mode of feeding or handling of prey. No obvious form of resource partitioning has been observed between the two species, although it may exist. The hottentot, Pachymetopon blochii, is the most common fish species in environments where janolids have been observed and is known to be a generalized omnivore of benthic organisms (Nepgen 1977). Specimens of Bonisa nakaza, Janolus capensis, and J. longidentatus are readily ingested by hottentots in the field and even more readily regurgitated in an unharmed state, suggesting defensive immunity to fish predation. More detailed study of the three species of janolids is required to determine their competitive population dynamics, interaction, and possible resource partitioning. ACKNOWLEDGEMENTS I should like to thank Dr Roberta Griffiths of the University of Cape Town for her critical review of the manuscript. Miss Patricia Cook of the British Museum (Natural History) kindly identified the ectoproct species. The follow- ing members of staff at the South African Museum provided valuable assistance in the preparation of the manuscript: Mr William R. Liltved collected many specimens and made many of the original drawings and all the final inked figures; Mr Sidney X. Kannemeyer and Mrs Michelle van der Merwe prepared the final photographic prints; Mrs Gail Strong typed several drafts of the manuscript, and Mrs Marcelle Scheiner typed the final draft. To them I extend my appreciation. SOUTH AFRICAN JANOLIDAE 41 REFERENCES ALDER, J. 1844. On the genus Venilia. Ann. Mag. nat. Hist. 13: 407. ALDER, J. & Hancock, A. 1844. Description of a new genus of nudibranchiate Mollusca, with some new species of Eolis. Ann. Mag. nat. Hist. 13: 161-167. ALDER, J. & Hancock, A. 1848. Additions to the British species of nudibranchiate Mollusca. Ann. Mag. nat. Hist. (2) 1: 189-192. ALDER, J. & Hancock, A. 1851. Monograph of the British Nudibranchiate Mollusca 5. London: Ray Society. ALDER, J. & Hancock, A. 1854a. Notice of some new species of British Nudibranchiata. Ann. Mag. nat. Hist. (2) 14: 102-105. ALDER, J. & Hancock, A. 18546. Monograph of the British Nudibranchiate Mollusca 6. Londou: Ray Society. ALDER, J. & Hancock, A. 1855. Monograph of the British Nudibranchiate Mollusca 7. London: Ray Society. BaBA, K. & ABE, T. 1970. Two new species of Janolidae from Toyama Bay, Japan (Gastro- poda: Nudibranchia). Veliger 13: 63-66. BARNARD, K. 1927. South African nudibranch Mollusca, with descriptions of new species, and a note on some specimens from Tristan d’ Acunha. Ann. S. Afr. Mus. 25: 171-215. BErGH, R. 1874. Beitrage zur Kenntniss der Aeolidiaden.2. Verh. zool-bot Ges. Wien 24: 395-416. BERGH, R. 1884. Report on the Nudibranchiata. Report of the Scientific Results of the Voyage of the H.M.S. Challenger, Zool. 10(26): 1-154. BERGH, R. 1888. Beitrage zur Kenntniss der Aeolidiaden, 9. Verh. zool-bot. Ges. Wien 38: 673-706. BeErGH, R. 1904. Malacologische Untersuchungen (9)6(1). Jn: SEMPER, C., ed. Reisen im Archipel der Philippinen. Wiesbaden: Kreidel’s Verlag. BERGH, R. 1907. The Opisthobranchiata of South Africa. Trans. S. Afr. phil. Soc. 17(1): 1-44. BoucueEt, P. 1975. Nudibranches nouveaux des cotes du Sénégal. Vie Milieu 25(1A): 119-131. Burn, R. 1958. Further Victorian Opisthobranchia. J. malac. Soc. Austr. 2: 20-36. Burn, R. & Miiier, M. 1969. A new genus, Caldukia, and an extended description of the type species, Proctonotus? affinis Burn, 1958 (Mollusca Gastropoda: Arminacea, Antiopelli- dae). J. malac. Soc. Austr. 12: 23-31. CHIAJE, S. 1841. Descrizione e notomia degli animali invertebrati della Sicilia citeriore osservati vivi negli anni 1822-1830, Napoli. 7: pls 86-173. CocKERELL, T. & Exior, C. 1905. Notes on a collection of Californian nudibranchs. J. malac. 12(3): 31-53. Cooper, J. 1863. On new or rare Mollusca inhabiting the coast of California. 2. Proc. Calif. Acad. Sci. 3: 56-60. Euror, C. 1906. Notes on some British nudibranchs. J. mar. biol. Ass. U.K. 7(3): 333-382. Euior, C. 1907. Nudibranchs from New Zealand and the Falkland Islands. Proc. malac. Soc. Lond. 7: 327-361. Eviot, C. 1909. Report on the nudibranchs collected by Mr. James Hornell at Okhamandal in Kattiawar in 1905-6. In: Report to the government of Baroda on the marine zoology of Okhamandal in Kattiawar 1: 137-145. Eviot, C. 1910. A monograph of the British nudibranch Mollusca 8 (suppl.). London: Ray Society. Franc, A. 1968. Mollusques gastéropodes et scaphopodes. Jn: Grassk, P., ed. Traité de Zoologie 5(3): Paris: Masson et Cie. Hoye, W. 1902. Two points in nomenclature. 1. Cypriniadea versus Cyprina. 2. The genus Antiopa. J. Conch., Lond. 10(7): 214. Hurst, A. 1967. The egg masses and veligers of thirty northeast Pacific opisthobranchs. Veliger 9: 255-288. Jounson, M. & Snook, H. 1927. Seashore animals of the Pacific Coast. New York: Dover. MACFaRLAND, F. 1966. Studies of opisthobranchiate mollusks of the Pacific coast of North America. Mem. Calif. Acad. Sci. 6: 1-546. MacGinitie, G. & MacGinitie, N. 1949. Natural history of marine animals. New York: McGraw-Hill. 42 ANNALS OF THE SOUTH AFRICAN MUSEUM Marcus, E. 1955. Opisthobranchia from Brazil. Bol. Fac. Fil. Cien. Let. Univ. S. Paulo, Brazil (Zool.) 20: 89-261. Marcus, E. 1958. On western Atlantic opisthobranchiate gastropods. Amer. Mus. Novit. 1906: 1-82. MiLLer, M. 1970. Two new species of the genus Caldukia Burn & Miller, 1969 (Mollusca: Gastropoda: Opisthobranchia) from New Zealand waters. Veliger 12: 279-289. Miter, M. 1971. A new genus and species of the nudibranch family Janolidae (Mollusca, Gastropoda, Opisthobranchia) from New Zealand waters. J. nat. Hist. 5: 491-498. NEPGEN, C. 1977. The biology of the hottentot, Pachymetopon blochii (Val.) and the silverfish Argyrozona argyrozona (Val.) along the Cape south-west coast. Invest. Rep. Sea Fish. Branch 105: 1-35. O’DOoNOGHUE, C. 1922. Notes on the taxonomy of nudibranchiate Mollusca from the Pacific Coast of North America. Proc. malac. Soc. Lond. 15: 133-150. O’DonoGHuE, C. 1924. Notes on the nudibranchiate Mollusca from the Vancouver Island Region. 4. Trans. R. Can. Inst. 15(1): 1-33. Pruvot-FoL, A. 1954. Mollusques opisthobranches. Faune Fr. 58: 1-460. QUATREFAGES, J. 1844. Sur les gastéropodes phlébentérés (Phlébentérata Nob.), ordre nouveau de la classe des gastéropodes, proposé d’aprés l’examen anatomique et physiologique des genres Zéphyrine (Zephyrina Nob.), Actéon (Acteon Oken), Actéonie (Acteonie Nob.), Amphorine (Amphorina Nob.), Pavois (Pelta Nob.), Chalide (Chalidis Nob.). Ann. Sci. Nat., Paris (3) Zool. 1: 129-183. Ro.ier, R. 1970. A list of recommended nomenclatural changes for MacFarland’s ‘Studies of Opisthobranchiate Mollusks of the Pacific Coast of North America’. Veliger 12: 371-374. SCHMEKEL, L. 1970. Anatomie der Genitalorgane von Nudibranchiern (Gastropoda Euthyneura). Pubbl. Staz. Zool. Napoli 38: 120-217. STEINBERG, J. 1963. Notes on the opisthobranchs of the West Coast of North America. III. Further nomenclatural changes in the order Nudibranchia. Veliger 6: 63-67. THIELE, J. 1931. Handbuch der Systematischen Weichterkunde 1. Jena: Gustav Fischer. THompson, T. & Brown, G. 1976. British opisthobranch molluscs. Synopses of the British fauna (New Series) No. 8. London: Academic Press. VERANY, G. 1845. Janus spinolae. Mag. Zool. (2) 7: 121-122. ABBREVIATIONS a anus 1 intestine ag anal glands m muscle bands al albumen gland me membrane gland am ampulla mu mucous gland b buccal mass n nephroproct be bursa copulatrix O oral glands Cc cerebral ganglion Pp penis cp cerebropleural ganglion pg pedal ganglion d digestive gland branch pl pleural ganglion dg digestive gland rc. renopericardial canal C eye rs receptaculum seminis f female gland mass S stomach g gonopores sg Salivary gland ga _ceratal ganglion vd ___-vas deferens gl ceratal gland 6. SYSTEMATIC papers must conform to the Jnternational code of zoological nomenclature (particularly Articles 22 and S51). Names of new taxa, combinations, synonyms, etc., when used for the first time, must be followed by the appropriate Latin (not English) abbreviation, e.g. gen. nov., sp. nov., comb. nov., Syn. nov., etc. An author’s name when cited must follow the name of the taxon without intervening punctuation and not be abbreviated; if the year is added, a comma must separate author’s name and year. The author’s name (and date, if cited) must be placed in parentheses if a species or subspecies is transferred from its original genus. The name of a subsequent user of a scientific name must be separated from the scientific name by a colon. Synonymy arrangement should be according to chronology of names, i.e. all published scientific names by which the species previously has been designated are listed in chronological order, with all references to that name following in chronological order, e.g.: Family Nuculanidae Nuculana (Lembulus) bicuspidata (Gould, 1845) Figs 14-15A Nucula (Leda) bicuspidata Gould, 1845: 37. Leda plicifera A. Adams, 1856: 50. Laeda bicuspidata Hanley, 1859: 118, pl. 228 (fig. 73). Sowerby, 1871: pl. 2 (fig. 8a—b). Nucula largillierti Philippi, i861: 87. Leda bicuspidata: Nicklés, 1950: 163, fig. 301; 1955: 110. Barnard, 1964: 234, figs 8-9. Note punctuation in the above example: comma separates author’s name and year “semicolon separates more than one reference by the same author full stop separates references by different authors figures of plates are enclosed in parentheses to distinguish them from text-figures dash, not comma, separates consecutive numbers Synonymy arrangement according to chronology of bibliographic references, whereby the year is placed in front of each entry, and the synonym repeated in full for each entry, is not acceptable. In describing new species, one specimen must be designated as the holotype; other speci- mens mentioned in the original description are to be designated paratypes; additional material not regarded as paratypes should be listed separately. The complete data (registration number, depository, description of specimen, locality, collector, date) of the holotype and paratypes must be recorded, e.g.: Holotype SAM-—A13535 in the South African Museum, Cape Town. Adult female from mid-tide region, King’s Beach Port Elizabeth (33°51’S 25°39’E), collected by A. Smith, 15 January 1973. : Note standard form of writing South African Museum registration numbers and date. 7. SPECIAL HOUSE RULES Capital initial letters (a) The Figures, Maps and Tables of the paper when referred to in the text *) e.g. ‘.. . the Figure depicting C. namacolus...’; ‘*.. . in C. namacolus (Fig. 10)... - (b) The prefixes of prefixed surnames in all languages, when used in the text, if not preceded by initials or full names e.g. DuToit but A.L.du Toit; Von Huene but F. von Huene (c) Scientific names, but not their vernacular derivatives e.g. Therocephalia, but therocephalian Punctuation should be loose, omitting all not strictly necessary Reference to the author should be expressed in the third person Roman numerals should be converted to arabic, except when forming part of the title of a book or article, such as ‘Revision of the Crustacea. Part VIII. The Amphipoda.’ nae Specific name must not stand alone, but be preceded by the generic name or its abbreviation to initial capital letter, provided the same generic name is used consecutively. Name of new genus or species is not to be included in the title: it should be included in the abstract, counter to Recommendation 23 of the Code, to meet the requirements of Biological Abstracts. TERRENCE M. GOSLINER THE SOUTH AFRICAN JANOLIDAE (MOLLUSCA, NUDIBRANCHIA) WITH THE DESCRIPTION OF A NEW GENUS AND TWO NEW SPECIES ea OF THE SOUTH AFRICAN CAPE TOWN _ INSTRUCTIONS TO AUTHORS 1. MATERIAL should be original and not published elsewhere, in whole or in part. 2. LAYOUT should be as follows: (a) Centred masthead to consist of Title: informative but concise, without abbreviations and not including the names of new genera or species Author’s(s’) name(s) Address(es) of author(s) (institution where work was carried out) Number of illustrations (figures, enumerated maps and tables, in this order) (b) Abstract of not more than 200 words, intelligible to the reader without reference to the text (c) Table of contents giving hierarchy of headings and subheadings (d) Introduction (e) Subject-matter of the paper, divided into sections to correspond with those given in table of contents (f) Summary, if paper is lengthy (g) Acknowledgements (h) References (i) Abbreviations, where these are numerous 3. MANUSCRIPT, to be submitted in triplicate, should be typewritten and neat, double spaced with 2,5 cm margins all round. First lines of paragraphs should be indented. Tables and a list of legends for illustrations should be typed separately, their positions indicated in the text. All pages should be numbered consecutively. Major headings of the paper are centred capitals; first subheadings are shouldered small capitals; second subheadings are shouldered italics; third subheadings are indented, shouldered italics. Further subdivisions should be avoided, as also enumeration (never roman numerals) of headings and abbreviations. Footnotes should be avoided unless they are short and essential. Only generic and specific names should be underlined to indicate italics; all other marking up should be left to editor and publisher. 4. ILLUSTRATIONS should be reducible to a size not exceeding 12 « 18 cm (19 cm including legend); the reduction or enlargement required should be indicated; originals larger than 35 x 47 cm should not be submitted; photographs should be rectangular in shape and final size. A metric scale should appear with all illustrations, otherwise magnification or reduction should be given in the legend; if the latter, then the final reduction or enlargement should be taken into consideration. All illustrations, whether line drawings or photographs, should be termed figures (plates are-not printed; half-tones will appear in their proper place in the text) and numbered in a single series. Items of composite figures should be designated by capital letters; lettering of figures is not set in type and should be in lower-case letters. The number of the figure should be lightly marked in pencil on the back of each illustration. 5. REFERENCES cited in text and synonymies should all be included in the list at the end of the paper, using the Harvard System (ibid., idem, loc. cit., op. cit. are not acceptable): (a) Author’s name and year of publication given in text, e.g.: ‘Smith (1969) describes .. .’ ‘Smith (1969: 36, fig. 16) describes...’ ‘As described (Smith 1969a, 1969b; Jones 1971)’ ‘As described (Haughton & Broom 1927)...’ ‘As described (Haughton et al. 1927)...’ Note: no comma separating name and year Dagination indicated by colon, not p. names of joint authors connected by ampersand - et al. in text for more than two joint authors, but names of all authors given in list of references. (b) Full references at the end of the paper, arranged alphabetically by names, chronologically within each name, with suffixes a, b, etc. to the year for more than one paper by the same author in that year, e.g. Smith (1969a, 19695) and not Smith (1969, 1969a). For books give title in italics, edition, volume number, place of publication, publisher. For journal article give title of article, title of journal in italics (abbreviated according to the World list o, scientific periodicals. 4th ed. London: Butterworths, 1963), series in parentheses, volume number, part number (only if independently paged) in parentheses, pagination (first and last pages of article). Examples (note capitalization and punctuation) BULLOUGH, W. S. 1960. Practical invertebrate anatomy. 2nd ed. London: Macmillan. FISCHER, P. —H. 1948. Données sur la résistance et de le vitalité des mollusques. J. Conch., Paris 88: 100-140. FiscHER, P.-H., DuvAL, M. & Rarry, A. 1933. Etudes sur les échanges respiratoires des littorines. Archs Zool. exp. zen. 74: 627-634. OH, A. J. 1960a. Ecological notes on Conus (Mollusca: Gastropoda) in the Trincomalee region of Ceylon. Ann. Mag. nat. Hist. (13) 2: 309-320. Konn, A. J. 19606. Spawning behaviour, ee masses and larval development in Conus from the Indian Ocean. Bull. Bingham oceanogr. Coll. 17 (4): THIELE, J. 1910. Mollusca: B. Polylsenphors Gastropoda marina, Bivalvia. In: SCHULTZE, L. Zoologische und anthropologische Ergebnisse einer Forschungsreise im westlichen und zentralen Siid-Afrika 4: 269-270. Jena: Fischer. Denkschr. med.-naturw. Ges. Jena 16: 269-270. (continued inside back cover) ANNALS OF THE SOUTH AFRICAN MUSEUM ANNALE VAN DIE SUID-AFRIKAANSE MUSEUM Volume 86 Band October 1981 Oktober Part 2 Deel DESCRIPTION OF SOME JUVENILE HOMINID SPECIMENS FROM SWARTKRANS, TRANSVAAL By bo Ee GRINE Cape Town Kaapstad The ANNALS OF THE SOUTH AFRICAN MUSEUM are issued in parts at irregular intervals as material becomes available Obtainable from the South African Museum, P.O. Box 61, Cape Town 8000 Die ANNALE VAN DIE SUID-AFRIKAANSE MUSEUM word uitgegee in dele op ongereelde tye na gelang van die beskikbaarheid van stof Verkrygbaar van die Suid-Afrikaanse Museum, Posbus 61, Kaapstad 8000 OUT OF PRINT/UIT DRUK ie ees) 53), LD, 6-59, hatin), ese, ED): GG, tat), WO). 8, TD, My, MOGES, GE Sy. I east, 1SCES), LO), 27, SIGE), WIG), 28, 250) EDITOR/REDAKTRISE Ione Rudner Copyright enquiries to the South African Museum Kopieregnavrae aan die Suid-Afrikaanse Museum ISBN 0 86813 024 9 Printed in South Africa by In Suid-Afrika gedruk deur The Rustica Press, Pty., Ltd., Die Rustica-pers, Edms., Bpk., Court Road, Wynberg, Cape Courtweg, Wynberg, Kaap DESCRIPTION OF SOME JUVENILE HOMINID SPECIMENS FROM SWARTKRANS, TRANSVAAL By F. E. GRINE South African Museum, Cape Town (With 15 figures) [MS accepted 12 May 1981] ABSTRACT Recent work on the hominid fossils from Swartkrans revealed several juvenile specimens which required cleaning and description. Four such specimens, representing the gnathic and/or dental remains of three ‘robust’ australopithecine individuals, have been prepared and described in detail. The specimens, SK 839/852, SK 1595, and SK 2147 have added a number of deciduous and permanent teeth to the collection from this site. CONTENTS PAGE HA tKOCUCEOM rere Cisse aac wea eee ee, 43 Matcnalandidescriplions (2.2 0 9.2. bees ie 45 SPECIMENISK 839 Wao eh ac Hos ee ak eee 45 WeciduousMdentitions. 4.4) ase ee eee eee 45 Rermanent Gentittion= 2445.) eae eee 50 SPECIMIEMESIKGS 592m eke ate es vd ore PS 53 Deciduousidentition= =. >.e0 ee see ee oo 54 Renmanentdentitionas ole) oe eee 58 Specimens SK 839/SK 852 composite ............ 61 SPECIMMEMYS RGIS OSes ges Reais noe omens a ie 62 Weciduousidentitions. 1.) vo sso ee ee ae 63 Rermanentadentitions oe) eee eee 65 Specimen!" SKQ14 7 ec oy ee see A 67 DISCUSSION Hye eee Oe tae chis es cpg AIOE 69 ANCKNOWIEdSeMIENES = hiss hairs bagels Hoe eee oe 69 FRETENEMCES ee ere ere ee eames eg i ae eee 70 INTRODUCTION Since 1948, when Robert Broom and J. T. Robinson began working at Swartkrans, the fossiliferous deposits of this cave system have yielded an abundance of hominid remains, including a number of juvenile specimens. Geological and palaeontological investigations have demonstrated that the Swartkrans deposit consists of older and younger sediment groups (Brain 1958, 1976; Cooke 1963, 1978; Brain, Vrba & Robinson 1974; Hendey 1974; Vrba 1975; Butzer 1976). Butzer (1976) and Brain (1976) have defined the older breccias as Member 1 and the younger breccias as Member 2 of the Swartkrans Formation. 43 Ann. S. Afr. Mus. 86 (2), 1981: 43-71, 15 figs. 44 ANNALS OF THE SOUTH AFRICAN MUSEUM The first hominid specimens recovered from the Swartkrans Formation were attributed to a new taxon, Paranthropus crassidens, by Broom (1949). The vast majority of hominid specimens which have been found at Swartkrans (over 95 % of individuals) have been referred to this taxon under a variety of names, viz. Paranthropus crassidens, P. robustus, Australopithecus crassidens, A. robustus, and A. robustus crassidens. Geological studies have shown that these australopi- thecine remains are derived from the Member 1 sediments (Brain 1976, 1978). The existence of. individuals of the genus Homo in the Swartkrans deposit has been well substantiated (Broom & Robinson 1949, 1950, 1952; Robinson 1953a, 1953b, 1961; Tobias 1968, 1978; Clarke, Howell & Brain 1970; Clarke & Howell 1972; Wallace 1972, 1975; Clarke 1977a, 1977b; Olson 1978). The first specimen referable to Homo to be found at Swartkrans was the mandible SK 15 (Broom & Robinson 1949, 1950). The jaw was found in a pocket of brownish breccia surrounded by the typically pink Member 1 sediments, and it was initially thought that SK 15 was younger than the australopithecine remains (Broom & Robinson 1949, 1950). Later, however, Robinson (1953b) consid- ered that the brown breccia encasing the mandible was a pocket of decalcified primary breccia and that SK 15 and the australopithecine remains were coeval. More recent work has demonstrated that the brown breccia is, indeed, younger than the australopithecine-bearing Member 1 sediments (Brain, Vrba & Robin- son 1974; Brain 1976, 1978). Three other hominid specimens, a P; (SK 18a), a proximal radius (SK 18b), and the buccal moiety of a P, (SK 43), were recovered from within the same pocket of Member 2 breccia as the SK 15 mandible. These four specimens have been referred to the same taxon (Broom & Robinson 1950, 1952; Robinson 19536; Tobias & Wells 1967). Several specimens attributed to Homo (e.g. SK 45, SK 847, SK 27, and SK 2635) have been recovered from the australopithecine-bearing Member 1 breccia. As noted by Clarke (1977a), there is little morphological basis for assuming that the Homo specimens from Members 1 and 2 belong to the same specific taxon. On the other hand, it is generally accepted that the australopi- thecine remains from Swartkrans are representive of a single, species-specific taxon (Broom & Robinson 1952; Robinson 1956; Tobias 1967; Wallace 1972; Clarke 1977a; Howell 1978). Recent work on the hominid fossils from Swartkrans revealed several juvenile specimens which required cleaning, reconstruction and description. Four such fossils are dealt with in this paper. The specimens consist generally of small gnathic fragments with variously preserved deciduous and permanent teeth. The fossils are catalogued in the Transvaal Museum as SK 839, SK 852, SK 1595, and SK 2147, and all are derived from Member 1 breccias. Each specimen will be described as follows: (i) the data presented on the relevant museum catalogue card, (ii) a brief statement of preservation prior to cleaning and reconstruction, and (iii) a detailed morphological description of the specimen after restoration. ‘JUVENILE HOMINID SPECIMENS FROM SWARTKRANS 45 All measurements recorded here were taken by the author, unless stated otherwise. The dental measurements were taken to 0,1 mm accuracy with a dial-equipped sliding vernier caliper. MATERIAL AND DESCRIPTIONS SPECIMEN SK 839 This specimen was excavated from Member 1 breccia by J. T. Robinson in 1952. The description on the catalogue card reads: ‘Fragmentary juvenile maxilla containing some deciduous and permanent teeth some of which have suffered considerable damage.’ Prior to restoration, this specimen consisted of isolated, incompletely developed and slightly damaged LM' and RM’, isolated and slightly damaged Rdm* and the distal two-thirds of the Rdm'. In addition, the badly damaged permanent incisors, the damaged Rdi', the root of the Ldi', and a small bit of the alveolar portion of the maxilla were preserved in a piece of breccia (Fig. 1). Fig. 1. Stereoview of the maxillary central deciduous and central and lateral permanent incisors of SK 839 prior to preparation. Compare with Figures 2-5. Scale in cm. The crown of the Ldi'! was found in a box with another, unrelated, hominid specimen from Swartkrans. The two deciduous central incisors and the four permanent incisors were cleaned, reconstructed and strengthened by the author. Deciduous dentition (Figs 2-4) Maxillary central deciduous incisors The damaged crowns and roots of both incisors are present. The crown of the Ldi’ is only moderately well preserved; the distolinqual quadrant is missing and the remaining lingual portion is separated from the buccal part by a narrow crack. The buccal surface is displaced somewhat by virtue of breaks on either side of a 0,7 mm wide vertical sliver of enamel. The crown of the Rdi’ is better preserved, with small pieces of both the mesial and distal edges missing. The roots of both teeth are poorly preserved; the lingual sides of both are missing just beyond the cervical lines, and the buccal surfaces are covered by remnants of the alveolar portion of the maxilla. 46 ANNALS OF THE SOUTH AFRICAN MUSEUM Occlusal wear is moderate in degree, with some reduction of crown height; most of the lingual aspect comprises a lingually sloping and slightly mesiodis- tally concave dentine exposure. The buccal edge of this surface is rather sharp and comprises a thin wall of enamel. Lingually, the remaining enamel surface shows slight faceting. Interproximal attrition is moderate on the mesial face, with dentine exposed on the incisal aspect of this surface; distally, interproxi- mal wear appears to be slightly less severe. Neither crown shows any periky- mata, but on the buccal surface of the Ldi’ there is a rather large hypoplastic area (MD, 1,0 mm; height, 0,6 mm) surrounded by several smaller pits, whilst on the Rdi' a considerably smaller, single hypoplastic pit is evident. The buccal aspect of the crown appears to have been square in outline before wear. There is no cervical enamel prominence, and this surface is nearly flat and disposed vertically incisocervically. Wear and damage to the incisolingual aspect of the crowns have obliterated any morphology that might have been present. Fig. 2. Lingual view of the maxillary central deciduous incisors; and occlusal view of the Rdm!' and Rdm/ of SK 839. Scale in cm. Robinson (1956: 122) recorded the existing mesiodistal diameters of both crowns as 5,7 mm and noted that, because of the appreciable degree of incisal and interproximal wear, the original dimensions ‘must have been about a millimetre greater’. The author agrees with Robinson that about 0,5 mm of dental material has been lost from either side of each crown. The dimensions of these teeth recorded by the present author are as follows: MD diameter MD diameter BL diameter (as measured) (estimated) (estimated) led. Le ee 5,7 mm 6,7 mm 5,2 mm Ridit syn eee 5,8 mm 6,8 mm 5,2 mm JUVENILE HOMINID SPECIMENS FROM SWARTKRANS Fig. 3. Oblique incisolingual view of the maxillary deciduous central incisors of SK 839, showing type of wear. Scale in cm. Fig. 4. Buccal view of maxillary central deciduous incisors and the attached alveolar bone of SK 839. Scale in cm. 47 48 ANNALS OF THE SOUTH AFRICAN MUSEUM Maxillary first deciduous molar The distal two-thirds or more of the right crown is preserved. The portion of the crown that is missing has been broken away along a line which runs obliquely buccolingually from a point apparently just behind the tip of the paracone, through the protocone to the level of the lingual developmental groove. A slight crack runs distally from the paracone, through the tip of the metacone where it bifurcates, one limb courses distobuccally, whilst the other divides the distal marginal ridge. The mesiobuccal root is missing, and the lingual root has been broken away from the crown with the cervical enamel margin. The distobuccal root is preserved for what appears to be most of its length. It is evident that buccally the neck of the radicular system was rather low, there being less than 1 mm of cementum exposed between the cervical enamel margin and the point of bifurcation of the two buccal roots. The distobuccal root, as preserved, slopes away from the crown both distally and very slightly buccally. The root is straight and the apical end appears to taper slightly. It is evident that all four principal cusps—paracone, metacone, protocone and hypocone—were present. The size of the paracone cannot be determined accurately, but it seems to have been approximately the same size as, or perhaps slightly smaller than, the metacone. The protocone was apparently the largest cusp. The hypocone and metacone are almost the same size. Occlusal wear is moderate and has produced two bevelled surfaces. Mesially, the protocone has been worn flat and a moderately large, concave dentine island is exposed. The protocone is worn slightly below the level of the paracone, which appears to have shown a very small dentine exposure. The distal portion of the occlusal surface slopes distolingually; a moderate-sized, concave dentine island is exposed on the hypocone, while no dentine is shown on the metacone. Interproximal attrition appears to have been slight to perhaps moderate in degree. The distal contact facet, which measures approximately 3,1 mm bucco- lingually by 2,1 mm in height, is ovoid in outline and flattened. It is located, for the most part, behind the hypocone, while the corresponding facet on the dm? is situated over the buccal half of its mesial face. The buccal face, as preserved, shows no cervical prominence or swelling, and it is faintly convex occlusocervically. There is no trace of a buccal groove. The cervical margin of the lingual face is not preserved, but this surface—over the hypocone—is more convex than the buccal surface. The lingual groove is very shallow and short. The protocone and paracone appear to have been separated by a very shallow, narrow groove. The trigon basin is represented by a short and narrow but oblique groove. The distal trigon crest is narrow but continuous. The distal marginal ridge is thick and high, extending without any apparent interruption from behind the tip of the metacone to the hypocone. The talon basin, or fovea posterior, is represented by a crescent-shaped, relatively deep groove. It is not possible to record any. crown diameters for this tooth. JUVENILE HOMINID SPECIMENS FROM SWARTKRANS 49 Maxillary second deciduous molar The crown of the right tooth is preserved. It is very nearly complete; the cervical margin has been damaged round the entire periphery save for the buccal aspect. A large, vertical sliver of enamel has broken away from the mesial face, and the distobuccal corner of the crown has been displaced slightly. Several narrow cracks traverse the crown. The radicular system has been broken away save for a short segment of the mesiobuccal root. Viewed from the occlusal aspect, the crown is square in outline. All four principal cusps are present and well developed. The protocone is by far the largest cusp, followed by the metacone and the paracone, the latter being slightly smaller. The hypocone is well developed and is approximately the same size as the metacone. Occlusal wear is slight, and has reduced the cuspal tips to nearly the same height. Wear is heaviest on the protocone, which shows large enamel facets but only a tiny pit of exposed dentine. Small dentine exposures are displayed also by the metacone and paracone. Generally, the mesial portion of the crown has been worn more heavily than the distal part. Interproximal contact with the dm’ appears to have been slight, while no distal contact facet is present. The mesial marginal ridge is thick and well developed and, though worn, it appears to have coursed without interruption from the front of the paracone to the protocone. Distal to the mesial marginal ridge, the fovea anterior is represented by a short, narrow groove which is completely enclosed distally by a short and thin but continuous epicrista. The epicrista runs parallel to the mesial marginal ridge. The trigon basin is broad and rather deep, its floor being represented by a triradiate groove. The distal trigon crest is moderately well developed but is incised and thinned in its middle by a narrow but deep groove. The distal marginal ridge is moderately thick and high and runs without interruption from the metacone to the hypocone. It is slightly lower in its middle, with two faint grooves on its mesial aspect; it runs continuously, high up on to the metacone. The talon basin is represented by a deep, broad groove which runs obliquely between the hypocone and protocone to incise the lingual surface of the crown. The buccal surface is rather flat and vertically disposed over the upper third of its height, and is slightly swollen and convex over the cervical two-thirds. There is no distinct cervical enamel prominence, and the cervical enamel line courses, for a short distance, towards the bifurcation of the two buccal roots. The buccal groove is rather weakly expressed; it is broad but very shallow towards its occlusal end, and courses cervically for less than half of the height of the crown where it terminates gradually. A number of very small hypoplastic pits cover the buccal face around both the occlusal and cervical extremities of the buccal groove. The lingual surface is more convex than the buccal, and curves in a continuous arc from the occlusal margin to the cervical line. The cervical enamel margin is missing, but it appears that no pronounced prominence was 50 ANNALS OF THE SOUTH AFRICAN MUSEUM present. The lingual groove is deep and broad over most of its course, but terminates gradually approximately half-way towards the cervical margin. There is no trace of a Carabelli trait. The protocone is ringed on its mesial and lingual aspects by a thin band of numerous, tiny, hypoplastic pits. This band is situated near the occlusal surface. The radicular system is represented solely by a short segment of the mesiobuccal root. In so far as it is preserved, it courses slightly mesialward and is considerably broader in its buccolingual than in its mesiodistal dimension. It appears that the radicular neck was low. Robinson (1956: 128) recorded the mesiodistal and buccolingual diameters of this crown as 10,5 mm and 11,4 mm respectively. The measurements obtained by the present author are as follows: MD diameter MDdiameter BL diameter (as measured) (estimated) (as measured) RGIS cet oe eee 10,7 mm 10,8 mm 11,5 mm Permanent dentition (Fig. 5) Maxillary central permanent incisors The left and right maxillary central incisors are represented. Both teeth consist of rather badly damaged crowns only. They are unerupted. A root is not present. The crowns were badly crushed and broken in preservation, and both have been reconstructed from several different pieces. The left crown is crushed and the central and distal regions of the lingual surface have been displaced. The mesial, buccal and distal surfaces have suffered considerable loss of enamel about the cervical margin. The mesiobuccal quadrant of the left crown is missing, and the mesial end of the lingual surface has suffered slight crushing. The mesial end of the incisal edge is slightly rounded, while the distal extremity of this edge shows a broad curvature. The incisal edge comprises large mesial and distal mammelons and a considerably smaller and lower median mammelon. Viewed from the buccal aspect, the crown is judged to have had a tapered outline, the cervical region having been narrower than the incisal portion. The buccal surface is gently convex incisocervically. Neither tooth exhibits either perikymatous or hypoplastic enamel. Lingually, there is a moderate cervical enamel swelling which appears to have been symmetrically disposed. The mesial marginal ridge is much more pronounced, and a relatively narrow, low median ridge projects vertically from the cervical enamel prominence towards the incisal edge. The lingual surface is slightly concave over the upper half of the crown, both incisocervically and mesiodistally. MD diameter MDdiameter BL diameter (as measured) (corrected) (corrected) TEL | ee oN sie 8,2 mm 8,6 mm = Ribaie. once 8,6 mm 8,6 mm 6,8 mm JUVENILE HOMINID SPECIMENS FROM SWARTKRANS al Maxillary lateral permanent incisors The crown of the right tooth is well preserved, with only slight damage to parts of the cervical enamel margin. The crown of the left tooth is severely crushed and distorted and, because the right crown is so well preserved, it was deemed unnecessary to attempt a reconstruction of its antimere. The tooth is unerupted. There is no root present. The crown appears to be completely developed. The mesial end of the incisal edge is gently rounded, while the distal extremity is slightly lower than the mesial and is more broadly rounded. The incisal edge is gently rounded and rather smooth; there is no mammelon development. Viewed from the buccal aspect, the crown has a tapered outline, with the cervical region narrower than the incisal portion. The buccal surface is slightly convex incisocervically, with a faint, broad, flattening in the centre of this surface over the upper half. This area is bounded both mesially and distally by faint ridges that course from the cervical region to the respective corners of the incisal edge. The distal side of this face displays several faint perikymatous ridges; there is no evidence of hypoplastic enamel. Lingually, the cervical region shows a slight swelling. The mesial marginal ridge is slightly developed, the distal marginal ridge is moderately well deve- loped and there is no median ridge present. The mesial and distal marginal ridges converge cervically to blend into the cervical swelling. The lingual surface is slightly concave, especially mesiodistally. MD diameter BL diameter Height (as measured) (as measured) (as measured) Rug hee). 6,0 mm 5,5 mm 8,7 mm Maxillary first permanent molars The unerupted crowns of both the left and right maxillary first permanent molars are present. Both crowns have suffered from damage to the cervical enamel; generally, the right tooth is better preserved. The buccal side of the left crown is separated and displaced from the rest of the tooth by a crack which runs through the tips of the paracone and metacone. Nothing of the radicular system is present. The two crowns are nearly identical in morphological detail (Fig. 5). Viewed from the occlusal aspect, the crown is nearly square in outline, with all four principal cusps present and well developed. The protocone is the largest cusp, followed by the paracone and metacone, which are nearly equal in size. The hypocone is the smallest cusp. The mesial marginal ridge is moderately thick; it is higher and thicker near the protocone where a moderate-sized protoconule is present. The protoconule is delineated on either side by shallow grooves. A short epicrista projects from the mesial end of the paracone and, although this crest is delineated on either side by deep, narrow grooves, there is SZ. ANNALS OF THE SOUTH AFRICAN MUSEUM Fig. 5. Lingual view of the permanent incisors and occlusal view of the first permanent molars of SK 839. Scale in cm. no anterior fovea present. The trigon basin is deep. The distal trigon crest is moderately thick, but is incised midway between the protocone and metacone by a relatively deep pit which is continuous with the deep, but narrow, groove between the protocone and hypocone. On the lingual surface this groove continues vertically for approximately half of the crown height, where it ends abruptly. The lingual surface projects from the occlusal to the cervical margin; the cervical half of this face is slightly convex. The mesiolingual corner of the protocone presents two short, vertical grooves which are separated by approxi- mately 2,7 mm of enamel. However, there is no alteration of the general curvature of this face. The Carabelli feature is represented by grooves. This feature is only slightly expressed on the left crown. The buccal surface is less expanded than the lingual and is only faintly convex occlusocervically. The buccal groove is vertically deep and narrow and extends only over the upper third of the crown. It ends abruptly. Neither the buccal nor the lingual surface shows any indication of hypoplastic enamel. Robinson (1956: 81) considered these crowns to be incompletely developed and recorded estimated, complete mesiodistal and buccolingual dimensions of 13,2 mm and 13,9 mm respectively. The present author agrees with Robinson that the crowns are immature, but he is unable to judge accurately how much, if any, additional increase in size would have been attained. The measurements of the crowns as recorded are as follows: MD diameter BL diameter (as measured) (as measured) |)" eaten os il eR pc 12,5 mm == RIVER ire el 12,8 mm 14,0 mm JUVENILE HOMINID SPECIMENS FROM SWARTKRANS Se) SPECIMEN SK 852 This specimen was excavated from Member 1 breccia by J. T. Robinson in 1952. The description on the catalogue card reads: ‘Poor specimen of juvenile mandible with left dm, (broken); dm,, moderately worn; right dm, appreciably worn; dm, moderately worn and incomplete; and fragment of M, erupting.’ Prior to cleaning and restoration, this specimen consisted of an isolated and nearly complete Rd, as well as the isolated, complete crowns of the LI,, LI, and LC. The major part of this specimen consisted of the poorly preserved mandibular corpora with the roots of the Ldm,, the damaged Rdm,, Ldm,, Rdm, and the metaconid of the RM,, and the matrix-covered crown of the LM, (Fig. 6). The mandibular corpora were preserved in misalignment, and wedged Fig. 6. Stereoview of the mandibular corpora and dentition of SK 852 prior to restoration. Compare with Figure 8. Scale in cm. between them was a large piece of unidentified bone. The region of the RM, was covered in plaster. The two corpora were removed from the matrix and cleaned. In this process the hypoconid and hypoconulid of the RM, were discovered beneath the plaster; this piece of crown was reconstructed in correct anatomical position. The corpora and their contained teeth were cleaned and, where necessary, strengthened with plaster. The mandibular corpora are poorly preserved, somewhat distorted and very incomplete. Meaningful descriptions or measurements for either of these are not possible (see Figs 8 and 10). 54 ANNALS OF THE SOUTH AFRICAN MUSEUM Deciduous dentition (Figs 7-9) Mandibular deciduous canine The right tooth only is present. It is reasonably well preserved with rather severe damage to the mesial portion of the crown, but the root is nearly complete with only the apical tip missing. Enamel from the mesial and mesiolingual parts of the crown has been broken away. The tip is damaged also and a moderately wide vertical crack courses across the middle of the buccal surface. Viewed from the buccal aspect, the crown is rather ‘mitten-shaped’ with a large primary cusp situated mesially and a small distal cuspulid. The buccal surface is faintly convex incisocervically over the cervical third of the crown’s height, and a cervical prominence is not present. The distal cuspulid is worn, but there remains the trace of a short, faint vertical groove between it and the main cusp. Fig. 7. Stereoview of the lingual aspect of the Rd. of SK 852. Scale in cm. Lingually, there is a slight cervical prominence which is skewed distally. There are no tubercles supported by this prominence. The cervical enamel line courses downwards from mesial to distal. The mesial third of this surface has been damaged. A moderately deep, triangular-shaped groove separates the main cusp from the distal cuspulid. A moderate enamel ridge courses from the tip of the distal cuspulid downwards to blend into the cervical enamel pro- minence. Occlusal wear is slight. A well-developed enamel facet is present along the distal slope of the main cusp and on the tip of the distal cuspulid. This facet is narrow but highly polished, and it is continuous along the main cusp to a horizontal disposition on the distal cuspulid. Dentine is not exposed. Interprox- imal attrition with the dm, was mild, with only a small, slightly flattened, JUVENILE HOMINID SPECIMENS FROM SWARTKRANS 5)5) circular contact facet (0,7 mm diameter) near the top of the distal cuspulid. The dimensions of the crown are as follows: MD diameter MD diameter BL diameter Height (as measured) (corrected) (as measured) (estimated) RG La 5,7 mm 5,9 mm 4,6 mm 6,5 mm The root of this tooth is single. It has a flattened ovoid outline in cross-section, with the longer axis (5,1 mm) running from mesiobuccal to distolingual and the shorter axis (3,8 mm) running from mesiolingual to disto- buccal. The root is rather straight and long. It is estimated that the original length was approximately 13,0 mm (measured length, 11,4 mm). Mandibular first deciduous molars The obliquely shorn-off roots of the left tooth and the somewhat damaged crown and roots of the right tooth are present. The right crown has suffered enamel loss along the entire distal surface, and the occlusal surface has a wide crack over the top of the hypoconid and metaconid. Two narrow, vertical cracks course across the buccal surfaces of the protoconid and hypoconid. Viewed from the occlusal aspect, the crown has a somewhat irregular rectangular outline. The mesial end is slightly narrowed and projects anteriorly, while the distal end of the crown is broad and appears to have been flattened. The protoconid, hypoconid, metaconid, and entoconid are preserved. It appears that the hypoconulid was absent, or if it was present originally it is evident that it would have been very small. Wear and damage do not permit an accurate assessment of the relative cuspal sizes. Occlusal wear is moderate, with considerable reduction and flattening of cuspal height. The occlusal surface has been worn so that a broad, rather flattened plane has been produced; this plane slopes slightly downward distally from the tips of the metaconid and protoconid. It appears that a small island of dentine was exposed on the top of the protoconid, while the metaconid shows a somewhat larger exposure. The top of the hypoconid is damaged, but it is evident that at least a moderately large dentine island was present. The entoconid is covered by a large, concave dentine basin. A second facet is present in front of the protoconid tip. This facet slopes downward mesially as a well-defined, polished enamel surface along the mesial marginal ridge. Interproximal wear is judged to have been mild both mesially, as discussed above, and distally. Although the distal surface of the dm, is missing, the mesial surface of the dm, presents a moderate-sized, rectangular and somewhat flattened contact facet. On the Ldm, this facet measures 3,5 mm in breadth and 2,3 mm in height. The mesial marginal ridge is relatively thick and courses round from in front of the protoconid to the mesial end of the metaconid. It is thicker buccally than lingually, where, mesial to the metaconid, it is considerably thinner. Nevertheless, the mesial marginal ridge is continuous around the front of the 56 ANNALS OF THE SOUTH AFRICAN MUSEUM tooth; it is not incised. Though the region is worn, it appears that a small plesioconulid might have been present mesial of the tip of the protoconid. The fovea anterior is represented by a short, relatively deep, transverse groove which is enclosed completely by the mesial marginal ridge and a continuous, high and moderately thick crest between the metaconid and protoconid. A broad contact is present between the metaconid and hypoconid, and the buccal groove is more mesial than the lingual groove. The lingual surface is slightly convex occlusocervically with a faint cervical enamel prominence present. The lingual groove is faint and fades imperceptibly approximately half-way down this face. The buccal surface over the hypoconid is rather bulbous in appearance and is moderately to markedly convex occlusocervically. A slight cervical enamel prominence is present. This face of the protoconid is convex over the cervical half, the occlusal half is somewhat flattened and slopes outwards cervically from the protoconid tip. A faintly developed tuberculum molare is present; the cervical portion of the protoconid surface projects slightly laterally and the cervical enamel line dips below the level of this margin on the hypoconid. The buccal groove is rather narrow, but relatively deep. It courses vertically to end abruptly about half-way down this surface. A small area of hypoplastic enamel is present on the buccal surface of the protoconid, just mesial to the buccal groove. The principal dimensions of the crown are as follows: MD diameter MD diameter BLtrigonid — BL talonid (as measured) (corrected) (as measured) (as measured) RGN eek) 4 ode 9,4 mm 10,0 mm 73) shoe 8,1 mm The radicular system comprises broad mesial and distal root plates and a very low neck. The two plates are directed almost vertically downward, expanding buccolingually towards their apical ends, and each has two separate radicular canals. The apical third of each root plate is bifid. The apical ends of the mesial and distal radiculae of the left tooth are separated on the buccal side by some 7,3 mm. The mesial root plate of the left tooth measures approximately 6,5 mm buccolingually at the cervical margin and 9,2 mm at the apical end. The distance from the cervical line to the buccal apex of the mesial plate is about 8,8 mm, and the lingual side of the plate measures some 9,3 mm in length. Mandibular second deciduous molars The left and right teeth are present. The left crown and root system are more complete than those of the right. The cervical half of the buccal surface and the upper part of the root of the left tooth are broken away. On the right tooth most of the protoconid and hypoconid have been broken away, together with much of the distal root plate. Both crowns show several fine cracks which course across the surfaces in various directions; the JUVENILE HOMINID SPECIMENS FROM SWARTKRANS 57 right tooth has been affected by this cracking more than the left. In so far as they are preserved, the crowns are nearly identical in morphological detail. Viewed from the occlusal aspect, the crown is rectangular in outline. The five principal cusps are present and well developed. The metaconid is by far the largest cusp with the hypoconid, protoconid and entoconid nearly equal in size. The hypoconulid is, by a slim margin, the smallest cusp. In general, the cusps are rather bulbous in appearance, with narrow and nearly vertically sided grooves separating them. Occlusal wear is slight, with all cusps showing enamel wear, though this is slightest on the entoconid. The other cusps have been reduced in height somewhat and they show flattened occlusal surfaces. The buccal cusps have been worn slightly more heavily than the lingual cusps. The tops of the protoconid, hypoconid, and hypoconulid are worn to a nearly flat, horizontal plane. Except for a small pit of dentine on the protoconid, there is no dentine exposure despite the reduction in cuspal height. Interproximal attrition with the dm,, as noted above, is slight, and there is no contact distally with the M,. The mesial marginal ridge of the dm, is very thick and, though worn, it appears to have been relatively high. It continues uninterrupted from in front of the metaconid across to the protoconid. Behind this ridge the fovea anterior is represented by an H-shaped groove. The mesial transverse groove is narrow and relatively shallow, and it is incompletely enclosed behind by short, access- ory transverse crests from the metaconid and protoconid. A short, narrow longitudinal groove separates these two crests. The distal transverse groove is somewhat longer, broader and deeper than the mesial groove, and it is completely enclosed behind by a very thick, continuous ridge between the back of the protoconid and the principal crest of the metaconid. The metaconid displays a broad contact with the hypoconid; the two buccal grooves, which surround the hypoconid, and the lingual groove, between the metaconid and entoconid, are arranged in a symmetrical Y-shaped pattern. The mesiobuccal developmental groove is thus situated mesiad of the level of the lingual groove. The distal marginal ridge is relatively thick and is uninterrupted in its course. This ridge is slightly worn but it is apparent that it supported a tiny accessory cuspulid (the tuberculum sextum) which is represented solely by a faint convex- ity of the distal crown surface. The cuspulid, despite its small size on the left tooth, is even smaller and more weakly expressed on the right crown. The fovea posterior is represented by a relatively deep but narrow crescent-shaped groove which is continuous with the talonid basin via a deep, narrow groove between the hypoconulid and entoconid. The buccal surface has suffered damage. However, it is evident that the mesiobuccal groove was better developed than the distobuccal groove. The distal groove is relatively deep but narrow, and extends vertically for about one-third of the crown height where it terminates rather abruptly. The lingual surface is somewhat inflated in appearance. It is slightly convex occlusocervically with a slight cervical enamel prominence at the base of the 58 ANNALS OF THE SOUTH AFRICAN MUSEUM metaconid and a rather marked prominence at the base of the entoconid. The enamel line continues round underneath the crown for a short distance between the mesial and distal radiculae. The lingual groove is of moderate depth but narrow, and continues vertically to the cervical margin. Over the lower third of its course it is bifid, with a faint accessory groove running parallel to it mesially. The distal half of the metaconid surface displays a number of minute hypoplas- tic pits, and these are more numerous and marked near the lingual groove. MD diameter BL trigonid BL talonid (as measured) (as measured) (as measured) Rai. 8h es 12,2 mm 9,7+ mm 9,.8+ mm RRC ier aces 12,4 mm _ — The buccolingual diameters recorded here are minimum values; because of the damage the buccal surface has suffered, it is not possible to estimate accurately the original dimensions of the trigonid and talonid portions of the crown. The root system of the second deciduous molar comprises broad mesial and distal radicular plates and a very low neck. Because of damage to the cervical region, the breadth of the plates cannot be measured accurately. The plates appear to diverge slightly as they course downward. The length of the buccal side of the mesial plate of the left tooth measures at least 10,5 mm. Permanent dentition (Figs 8-9) Mandibular central permanent incisor The isolated crown of the left central incisor is present. The crown is complete and well preserved. A very short segment of the developing root is represented on the mesial and distal aspects of the tooth. The tooth had not yet erupted at the time of death. The mesial and distal corners of the incisal edge are slightly rounded. A large, high mammelon is supported at either extremity, with a much lower tubercle between the mesial and distal mammelons. Viewed from the buccal aspect, the crown has a high, tapered outline; the mesiodistal diameter is considerably greater incisally than cervically. The buccal surface is slightly convex incisocervically and a cervical enamel prominence is not present. Numerous faint, horizontal, perikymatous lines cover this face entirely. Lingually, a moderate basal prominence is present. A tubercular structure is not supported by this swollen base. The basal prominence is symmetrically orientated. The mesial marginal ridge is thin and very faintly expressed; the distal marginal ridge is slightly better developed. The lingual surface is flattened. MD diameter BL diameter Height (as measured) (as measured) (as measured) Bee oeeeut 5,1 mm 5,7 mm 10,0 mm JUVENILE HOMINID SPECIMENS FROM SWARTKRANS 59 Mandibular lateral permanent incisor The isolated crown oi the leit tooth is present. The crown is well preserved and nearly complete, with only slight enamel loss to the buccal cervical and somewhat more severe damage to the base of the lingual side. The developing root is represented by short segments on the mesial and distal aspects of the tooth. The tooth had not yet erupted at the time of death. The crown is somewhat more robust in appearance than that of the central incisor. The mesial and distal comers of the incisal edge appear rather angulated with the mammelons at these extremities. The incisal edge is, mn general, horizontal, and it supports five small mammelons. Viewed from the buccal aspect, the crown is somewhat rectangular outline. The mesiodistal breadth across the incisal portion is greater than that = Fis. 9. Lingual view of the LC, LL, LI, and Rd, (from left to nght) of SK 852. Scale m cm. 60 ANNALS OF THE SOUTH AFRICAN MUSEUM across the cervical region of the crown, but the relative difference between these measurements is not so marked as in the central incisor. The buccal surface is shghtly convex incisocervically. Very faint perikymatous lines and a broad but shallow groove are present on this face. Although the cervical extremity of the lingual surface has been damaged, it is evident that a shght to moderate basal prominence was present. The basal swelling does not show tubercles. The mesial marginal ridge is slightly devel- oped; the distal marginal ridge is more faintly expressed than the mesial. Between these mdges, the lingual surface is faintly concave both mesiodistally and incisocervically. MD diameter BL diameter Height (as measured) (corrected) (as measured) | 3) SER es See nr 5,6 mm 6.4 mm 9.9 mm Mandibular permanent canine The isolated crown of the left mandibular canine is well preserved and is nearly complete. The lingual cervical enamel margin has suffered only shghtly from damage. Mesially, a short, thin sheet of the developing root is present, but this is not represented on any other part of the tooth. The tooth had not yet erupted at the time of death. Viewed from the buccal aspect, the crown is nearly square in appearance; the mesial corner of the tip is shghtly rounded, while the distal portion slopes downward for about half of the crown’s height from just behind the centre of the crown. The tip of the tooth is obtuse; mesially this surface is nearly horizontal. The buccal surtace is slightly convex occlusocervically as well as mesiodistally. No evidence of enamel pathology is shown on this face. Lingually, the cervical prominence is moderate and symmetrical. Lingual tubercules are not present. The mesial marginal mdge is faintly expressed; it appears as a low, thin band coursing round from the occlusal surface of the crown for approximately half of the crown height, where it blends impercept- ibly with the basal swelling. The distal marginal mdge is moderate; it originates occlusally approximately in the middle of the distal slope of the crown. This ridge becomes thickened towards the cervical prominence. A thin, low, vertical enamel ridge is situated mesiad of the distal marginal mdge, and this vertical crest originates at the cervical edge and continues cervically for just less than half of the crown height. The ‘median’ and distal marginal mdges are separated by a depression which is broader and relatively shallow occlusally, but which becomes increasingly narrower and deeper as it passes cervically, where it partially separates the distal mdge from the cervical swelling. The lingual surface is, for the most part, rather flat. MD diameter BL diameter Height (as measured) (as measured) (as measured) LC 7,3 mm 8.1 mm 9.0 mm ID SPSOMENS FROM SWARTIERANS 61 ees li ss stwated m &S opt and_ except for the occlusal surface. oaly the distal surface and the posterior half of the buccal face are exposed. A narrow crack courses across the tp of the hhypocomsuial and conimucs veriically over the desta] surface: sheht enamel dsplaccment has ocomred along tims cack. The root & represented by 2 relatively shori shoei below the cervical marem. and @ s&s evident that the mesial amd dssial rooi plaics had begun development at ai the time of death. The mght tooth i represented by a portion of the metacomd and a second picce of the the nehi Gown iS misses. Viewed irom the occlusal aspect. the Gown appears to have a rectansular outime. All five prmopal cusps ate presemi and are well developed The mctacomd scoms 10 be the largest cusp. Fe i svc by the protocomd. entocond, hypocomsd. and hypoconulad. — io have been moderately well developed. thick and comimuous. Distal to 1, the fowea amfcnmor i represenicad by a relatively shori_ bul Geep_ tamsverse sroove. The fowea SS dramed disially by a very narrow, but deep, stoove between the protocom”d and metacom. An “accessory amiemor fowea, 2s presented by the dm... t noi shown by tus tooth The prmopal cesi of the metacomd has a shorn. drstally deflected mdse (a Geiiecime wonkle) which contacts the hypoco- ma. Dhe meiacomd s more Gsially cloneaie than the protocomd. The imezual developmental sroove is situated Gistad to the mesiobuccal sroove—and thus the pumiary occlusal fissure paiiem ss symmetmally Y-shaped. The dsl maremal midec rs represenied by a small cospuhd (ihe tobercolum sexium) Winch ss demarcated from the hypoconulhd and eniocomd by shallow. narrow grooves. The fovea posicnor ts m the fomm of a small bat relatively Geep pit A Ralrow sroowe beiween the cntocomd and hypocomulkd commects the fovea postenor with the disiobuoccal sroove. The postenor part of the buccal surface ss shehily comvex occlesocerwacally- amd the dssiobuccal sroove s deep bat short. Ii courses vertically for siehily more than a third of the cown heieht, where i ends m a pi. The Gamensions of this cown cannot be deiemmmed accurately. sPnomens SK 8§39/SK 852 comPosiiz The niehi first and sccond mauillary Geadsous molars of SK $59 and the aie st. ne ee lanonships beiween bla ts fcat eked ac ivcry mood (Fie: 10). and they are very smmular m them sencral state of wear. Casts were made of the upper and lower scoond by Gnine (1977). The occlusal surfaces of the casis were exammed by scammme 62 ANNALS OF THE SOUTH AFRICAN MUSEUM Fig. 10. Lingual view of the occlusal relationship between the Rdm? of SK 839 and the Rdm, of SK 852. Scale in cm. electron microscopy. The types, degree, and orientation of the abrasive scrat- ches on the two teeth were found to be very similar. Also, the permanent teeth of the two specimens display degrees of calci- fication that are compatible with their having belonged to the same individual. The states of preservation of these fossils, for example the manganese staining patterns of the teeth, are also similar. These findings support Wallace’s (1972, 1973, 1978) suggestion that SK 839 and SK 852 represent the maxillary and mandibular remains of a single juvenile individual. SPECIMEN SK 1595 This specimen was recovered from the Member 1 breccia during the 1948-52 operations at Swartkrans. It was discovered in 1966 amongst a number of supposedly non-hominid faunal remains by C. K. Brain, who performed some preliminary preparation on it. The description on the catalogue card reads, ‘Fragment of maxilla with a slightly worn M' left and an erupting incisor?’ Prior to cleaning and restoration, this specimen consisted of a damaged left upper molar, the tip of a permanent lateral incisor and a small piece of what appeared to be the incisal edge of an incisor (Fig. 11). Also, a narrow, thin sheet of maxillary alveolar bone was present round the incisors, and at the back of the specimen the outline of what appeared to be part of an upper molar crown was visible in the matrix. Preparation revealed a very badly damaged and fragmented LI' (Figs 12-13), much of the crown of the LI? (Fig. 12), the mesial part of what is here considered to be the LM! (Fig. 14), and the distal moiety of what is here believed to be a slightly worn Ldm? (Fig. 14). JUVENILE HOMINID SPECIMENS FROM SWARTKRANS 63 Fig. 11. Stereoview of SK 1595 prior to preparation. Compare with Figures 12-14. Scale in cm. a = all Fig. 12. Lingual view of LI’ and LI* of SK 1595. Scale in cm. Deciduous dentition (Fig. 14) Maxillary second deciduous molar The distal half or more of the crown of the left second deciduous molar is present. The mesial portion of the crown has been broken away along an uneven transverse line which courses through the middle of the protocone lingually, across the distal edge of the paracone buccally. Nothing of the radicular system remains. It is evident that all four principal cusps were present originally, and it is probable that all were well developed. The metacone and hypocone are both well developed and they are nearly equal in size. Occlusal wear is slight on the preserved part of the crown. The protocone is the most heavily worn cusp; it has been reduced in height more than the two distal cusps, and it shows a large, faintly convex enamel facet. The hypocone 64 ANNALS OF THE SOUTH AFRICAN MUSEUM Fig. 13. Buccal view of LI’ of SK 1595. Scale in cm. Fig. 14. Occlusal view of the Ldm? and the mesial portion of the LM’ of SK 1595. Scale in cm. JUVENILE HOMINID SPECIMENS FROM SWARTKRANS 65 displays a large, flattened enamel facet which slopes downwards mesiobuccally. The metacone shows the lightest wear of the three cusps, with only a small enamel facet. Dentine is not exposed. An interproximal attrition facet is not present on the distal surface of the crown. Occlusally, the distal trigon crest is moderately thick but it is incised on its mesial and distal surfaces by a shallow groove. The distal marginal ridge is thick and complete. It is somewhat lower in its middle but is continuous high on to the distal aspect of the metacone. Four small cuspules are supported by this ridge; each cuspule is only faintly demarcated. The talon basin is relatively large and deep. A deep, narrow, longitudinal groove drains this basin near the middle of its mesial aspect, and this groove is continuous with the deep, narrow lingual groove between the protocone and hypocone. The lingual surface over the hypocone is moderately convex occlusocervi- cally and, although the cervical margin is damaged, it appears that a cervical enamel prominence was present. The lingual groove is relatively deep but narrow, and it courses vertically to the cervical margin. Approximately two- thirds of the way down the crown the groove is interrupted by a narrow (1,5 mm thick), horizontal enamel band. Buccally, the surface is more swollen in appearance than the lingual. The face of the metacone is moderately convex occlusocervically. The buccal groove is rather deep and somewhat broader than the lingual. It continues vertically for about half of the crown height, where it ends abruptly in a deep pit. The pit is surrounded buccally by a thin, continuous enamel crest from the paratone to the metacone. Near the distal side of the metacone a short, shallow and narrow, vertical groove courses from the occlusal margin. This groove partially demarcates the most buccal of the distal cuspule (= distostyle) from the main body of the metacone. A few minute hypoplastic pits are shown on the mesiobuccal surface of the metacone, near the occlusal margin of this surface. Some hypoplastic mottling is present also on the distal surface of the crown. The mesiodistal diameter of this crown is estimated to have been between 11,2 and 11,4 mm. The buccolingual diameter measured across the distal parts of paracone and protocone is some 12,5 mm, the original dimension is judged to have been perhaps another 0,2 mm greater. MD diameter BL diameter BL diameter (estimated) (as measured) (estimated) WdmMe res fa es. 11,3 mm 12,5 mm 12,7 mm Permanent dentition (Figs 12-14) Maxillary central permanent incisor The incomplete, isolated crown of the left maxillary incisor is represented. The crown, as preserved, was reconstructed from seven different fragments. Most of the lingual surface is missing, save for the mesial marginal ridge and a narrow bit of the incisodistal edge. The mesial surface is nearly complete, but 66 ANNALS OF THE SOUTH AFRICAN MUSEUM the cervical region is damaged. The buccal face is only slightly more complete than the lingual, where the mesial and incisal parts are present. The crown is unworn and probably had not yet erupted—at least it had not yet reached occlusion—at the time of death. The mesial corner of the incisal edge is squared off, with the incisal and mesial edges meeting at just slightly more than 90°. The distal corner is considerably more rounded than the mesial. The incisal edge appears gently curved mesiodistally. This edge is faintly scalloped, and the mesial part sup- ports two broad, low mammelons. Viewed from the buccal aspect, the crown appears to have been rather square in outline; its mesiodistal diameter is judged to have been only slightly less cervically than incisally. The buccal face is slightly convex incisocervically. Fine perikymatous lines are visible over much of the preserved surface. Lingually, the mesial marginal ridge is only slightly developed. It is extremely thin and low incisally, and expands somewhat as it courses cervically. MD diameter BL diameter Height (as measured) |S) a See Poet oes 9.0 mm = — Maxillary lateral permanent incisor The left maxillary lateral incisor is represented by an isolated, somewhat damaged crown. The cervical region round the entire periphery of the crown has suffered enamel loss, this being heaviest distobuccally and lingually. The crown appears to be fully, or nearly fully, developed, and it is relatively small, especially when compared to the I’. The tip of the crown is surmounted by two small mammelon-like tubercles. Mesial to the tip, the occlusal edge courses only slightly downward to the mesial corner, which is rather strongly curved. The edge distal to the tubercles slopes cervically much more strongly, and the distal ‘corner’ is broadly rounded and lower than the mesial. i Viewed from the buccal aspect, the crown is almost square in general outline, but with the distal occlusal ‘corner’ reduced. The buccal surface is faintly convex occlusocervically and there is no evidence of either perikymatous or hypoplastic enamel. Lingually, the cervical region is damaged; the presence, or otherwise, of the basal prominence cannot be determined. There is no evidence of the existence of lingual basal tubercles. Both the mesial and distal marginal ridges are faintly developed; they course round the periphery of the lingual face from the tip as thin, low enamel ridges. Towards the cervical margin, however, these ridges become slightly more prominent. The lingual surface is flat. MD diameter BL diameter Height (as measured) [es eae Ae 6,2 mm _ a JUVENILE HOMINID SPECIMENS FROM SWARTKRANS 67 Maxillary first permanent molar The left molar is represented by the damaged mesial portion of the crown. This piece consists of much of the mesial surface, the lingual part of the paracone, the entire mesial marginal ridge, and the mesial part of the paracone. The preserved part of the crown is unworn and the tooth is judged to have been unerupted at the time of death. The mesial marginal ridge is moderately thick and high. It is continuous, but is incised by two shallow grooves at the base of the tip of the paracone. A small cuspule (= epiconule) is demarcated by these grooves. The crest is expanded distally in its middle, and a small pit is situated atop it in this position. The fovea anterior is represented by an elongate, irregular and narrow transverse groove; it is broadest in front of the paracone. The proto- cone and paracone are separated by a relatively deep, broad groove by which the fovea anterior appears to be continuous with the trigon basin. Measurements are not possible. SPECIMEN SK 2147 (Fig. 15) There is no catalogue card for this specimen. It was discovered in the collection, labelled as an isolated lower premolar; it is, in fact, an isolated, incomplete Ldm'. Prior to preparation, the crown sat atop, and was partially covered by, a small piece of typical ‘Pink’, or Member 1, Swartkrans breccia. The specimen consists of the mesial two-thirds of a moderately worn crown, with much of the mesiobuccal root present. The distal part of the crown has been broken away along an irregular transverse line which cuts through the distal edge of the protocone, and through the lingual extent of the metacone where it turns sharply mesially to the posterior extent of the paracone. The crack extends directly lingually from this point along the plane occupied by the lingual developmental groove. Several narrow cracks traverse the protocone and the paracone. The cervical portion of the buccal surface is cracked, with considerable displacement of two rather large pieces of enamel. The mesial face of the crown has suffered slight enamel loss and cracking. A narrow horizontal crack traverses the mesial surface of the mesiobuccal root, and the tip of the lingual part of this root has been damaged near its apex. The occlusal outline of the crown cannot be determined. The protocone is a very large cusp; the paracone is well developed also, but considerably smaller than the protocone. Mesial to the paracone a large mesiostyle (= parastyle or paraconule) is present. The mesiostyle is separated from the paracone by a relatively deep, broad groove. Occlusal wear is moderate. The protocone has been reduced considerably in height, with a large, nearly flat and slightly lingually sloping enamel facet. A large, elongate and ovoid-shaped, concave dentine exposure is present in the middle of the protocone. The buccal side of the crown is less heavily worn; the 68 ANNALS OF THE SOUTH AFRICAN MUSEUM paracone, mesiostyle and mesial marginal ridge all show enamel wear. A small island of dentine is exposed on the tip of the paracone. Interproximal attrition, at least mesially, appears to have been moderate. A large, slightly concave facet for the d° is present at the buccal extremity of the mesial surface, it measures 2,8 mm buccolingually and 2,5 mm in height. Occlusally (Fig. 15), the mesial marginal ridge is well developed. It is thick and relatively high, and extends continuously from the mesiostyle to the mesiolingual aspect of the protocone. The fovea anterior is represented by a Fig. 15. Occlusal view of SK 2147 Ldm’. Scale in cm. relatively deep, broad groove which is continuous with the incision between the paracone and mesiostyle. The fovea is enclosed distally by a high, moderately thick enamel crest between the anterior part of the paracone and the proto- cone. The trigon basin is represented by a relatively deep and broad \-shaped groove. The tines of the A encompass part of the buccal aspect of the protocone between this cusp and the paracone and metacone. The stem of the \ has been damaged, but it is evident that it represented the buccal developmental groove between the paracone and metacone. The lingual surface of the protocone is slightly convex occlusocervically and a slight cervical enamel prominence is shown. The cervical portion of the buccal surface has been damaged, but this portion was seemingly less convex than the lingual face, with a slight cervical enamel prominence. The occlusal two-thirds of the buccal face is relatively flat. The buccal groove is missing. A faint groove courses vertically for a short distance from the occlusal incision between the paracone and the mesiostyle. It is apparent that there was no tuberculum molare over the mesiocervical aspect of the buccal surface. JUVENILE HOMINID SPECIMENS FROM SWARTKRANS 69 MD diameter BL diameter BL diameter (as measured) (corrected) Lich eee see sss 9.8 mm 10,0 mm The mesiobuccal root has a flattened, ovoid outline in cross-section; it is expanded buccolingually and compressed mesiodistally. The long axis of the root, in cross-section, runs slightly mesiobuccally-distolingually. The bucco- lingual diameter measures some 5,3 mm and the mesiodistal some 3,9 mm at the cervical margin. The root is relatively straight, and it courses rather markedly mesially from the cervical line to its tip. It diverges buccally some- what. This root is estimated to have been approximately 8,5 mm long. DISCUSSION Four specimens representing the jaws and/or teeth of some three juvenile hominid individuals from Swartkrans have been prepared and described in detail. The remains are all derived from Member 1 breccia. These specimens have added a number of deciduous and permanent teeth to the already sizeable collection from this site. Robinson (1956) referred one of these specimens (SK 839) to Paranthro- pus robustus. Wallace (1972) suggested that SK 839 and SK 852 represent a single individual, and he included this composite in his hypodigm of Australopi- thecus robustus from Swartkrans. The results of the present study support Wallace’s suggestion that SK 839 and SK 852 belonged to a single individual. SK 1595 was also referred to A. robustus by Wallace (1972) on the basis of the occlusal wear pattern exhibited by the dm’. The fourth specimen, SK 2147, was catalogued as an isolated permanent premolar. The author considers this tooth to be a left dm’, and as such it represents, to date, the second specimen of this tooth found at Swartkrans. The other dm’, SK 91, has been referred by both Robinson (1956) and Wallace (1972) to the ‘robust’ australopithecine taxon. The morphology shown by SK 2147 is similar to that evinced by SK 91, and the dm’ of SK 2147 is also referred to the Swartkrans ‘robust’ australopithecine taxon. An analysis of the morphological and metrical data available for the collection of deciduous and permanent teeth from Swartkrans is beyond the scope of the present paper and will be presented elsewhere. ACKNOWLEDGEMENTS I am grateful to Drs C. K. Brain and E. S. Vrba (Transvaal Museum) for permission to prepare and describe specimens in their care. I thank Prof. P. V. Tobias and Dr T. D. White for reading and constructively criticizing this paper. Mr A. R. Hughes and Mr H. Thackwray assisted with photography. This work was supported by a grant from the Senate Research Committee, University of the Witwatersrand. 70 ANNALS OF THE SOUTH AFRICAN MUSEUM REFERENCES BRAIN, C. K. 1958. The Transvaal ape-man bearing cave deposits. Transv. Mus. Mem. 11: 1-125. Brain, C. K. 1976. A re-interpretation of the Swartkrans site and its remains. S. Afr. J. Sci. 72: 141-146. BRAIN, C. K. 1978. Some aspects of the South African australopithecine sites and their bone accumulations. Jn: JoLty, C. ed. Early hominids of Africa: 131-161. London: Duckworth. BRAIN, C. K., VrBA, E. S. & RoBINson, J. T. 1974. A new hominid innominate bone from Swartkrans. Ann. Transv. Mus. 29: 55-63. Broom, R. 1949. Another new type of fossil ape-man (Paranthropus crassidens). Nature, Lond. 163: 57. Broom, R. & Rosinson, J. T. 1949. A new type of fossil man. Nature, Lond. 164: 322-323. Broom, R. & Rosinson, J. T. 1950. Man contemporaneous with the Swartkrans ape-man. Am. J. phys. Anthrop. 8: 151-156. Broom, R. & Rosinson, J. T. 1952. Swartkrans ape-man Paranthropus crassidens. Transv. Mus. Mem. 6: 1-123. Burzer, K. W. 1976. Lithostratigraphy of the Swartkrans Formation. S$. Afr. J. Sci. 72: 136-141. CLARKE, R. J. 1977a. The cranium of the Swartkrans hominid SK 847 and its relevance to human origins. Unpublished Ph.D. Thesis, University of the Witwatersrand, Johannes- burg. CLARKE, R. J. 1977b. A juvenile cranium and some adult teeth of early Homo from Swart- krans, Transvaal. S. Afr. J. Sci. 73: 46-49. CLARKE, R. J. & HowE LL, F. C. 1972. Affinities of the Swartkrans 847 hominid cranium. Am. J. phys. Anthrop. 37: 319-336. CLARKE, R. J., HOWELL, F. C. & Brain, C. K. 1970. More evidence of an advanced hominid at Swartkrans. Nature, Lond. 225: 1219-1222. Cooke, H. B. S. 1963. Pleistocene mammal faunas of Africa. Viking Fd Publ. Anthrop. 36: 65-116. CookE, H. B. S. 1978. Faunal evidence for the biotic setting of early African hominids. In: JoLiy, C., ed. Early hominids of Africa: 267-281. London: Duckworth. GrinE, F. E. 1977. Analysis of early hominid deciduous molar wear by scanning electron microscopy: a preliminary report. Proc. electron microsc. Soc. sth. Afr. 7: 157-158. HENDEY, Q. B. 1974. Faunal dating of the late Cenozoic of southern Africa. Quaternary Res., N.Y. 4: 149-161. HowELL, F. C. 1978. Hominidae. In: MaGuio, V. J. & Cooke, H. B.S., eds. Evolution of African mammals: 154-248. London: Harvard University. Otson, T. R. 1978. Hominid phylogenetics and the existence of Homo in Member 1 of the Swartkrans Formation, South Africa. J. hum. Evol. 7: 159-178. RoBINnson, J. T. 1953a. The nature of Telanthropus capensis. Nature, Lond. 171: 33. Rosinson, J. T. 1953b. Telanthropus and its phylogenetic significance. Am. J. phys. Anthrop. 11: 445-501. Rosinson, J. T. 1956. The dentition of the Australopithecinae. Transv. Mus. Mem. 9: 1-179. Rosinson, J. T. 1961. The australopithecines and their bearing on the origin of man and of stone-tool making. S. Afr. J. Sci. 57: 3-13. TosiAs, P. V. 1967. The cranium and maxillary dentition of Australopithecus (Zinjanthropus) boisei 2. Olduvai Gorge. Cambridge: Cambridge University. Tosias, P. V. 1968. Middle and early Upper Pleistocene members of the genus Homo in Africa. In: Kurtu, G. ed., Evolution and Hominisation: 176-194. Stuttgart: Fischer. TosiAs, P. V. 1978. The earliest Transvaal members of the genus Homo with another look at some problems of hominid taxonomy and systematics. Z. Morph. Anthrop. 69: 225-265. Tosias, P. V. & WeLLs, L. H. 1967. South Africa. In: OAKLEY, K. & CAMPBELL, B. eds. Catalogue of fossil hominids: 49-100. London: British Museum (Natural History). VrBA, E. S. 1975. Some evidence of chronology and palaeoecology of Sterkfontein, Swartkrans and Kromdraai from the fossil Bovidae. Nature, Lond. 254: 301-304. JUVENILE HOMINID SPECIMENS FROM SWARTKRANS TA WaLLAceE, J. A. 1972. The dentition of the South African early hominids: a study of form and function. Unpublished Ph.D. Thesis, University of the Witwatersrand, Johannesburg. WaLtace, J. A. 1973. Molar occlusion in the ape-man (Australopithecus). Am. J. Orthodont. 63: 606-609. WALLACE, J. A. 1975. Dietary adaptations of Australopithecus and early Homo. In: TuTTLE, R., ed. Paleoanthropology, morphology and paleoecology: 203-223. The Hague: Mouton. WALLACE, J. A. 1978. Evolutionary trends in the early hominid dentition. In: JoLiy, C. ed. Early hominids of Africa. London: Duckworth. a 6. SYSTEMATIC papers must conform to the /nternational code of zoological nomenclature (particularly Articles 22 and 51). Names of new taxa, combinations, synonyms, etc., when used for the first time, must be followed by the appropriate Latin (not English) abbreviation, e.g. gen. nov., sp. nov., comb. nov., syn. nov., etc. An author’s name when cited must follow the name of the taxon without intervening punctuation and not be abbreviated; if the year is added, a comma must separate author’s name and year. The author’s name (and date, if cited) must be placed in parentheses if a species or subspecies is transferred from its original genus. The name of a subsequent user of a scientific name must be separated from the scientific name by a colon. Synonymy arrangement should be according to chronology of names, i.e. all published scientific names by which the species previously has been designated are listed in chronological order, with all references to that name following in chronological order, e.g.: Family Nuculanidae Nuculana (Lembulus) bicuspidata (Gould, 1845) Figs 14-15A Nucula (Leda) bicuspidata Gould, 1845: 37. Leda plicifera A. Adams, 1856: 50. Laeda bicuspidata Hanley, 1859: 118, pl. 228 (fig. 73). Sowerby, 1871: pl. 2 (fig. 8a—b). Nucula largillierti Philippi, 1861: 87. Leda bicuspidata: Nicklés, 1950: 163, fig. 301; 1955: 110. Barnard, 1964: 234, figs 8-9. Note punctuation in the above example: comma separates author’s name and year “semicolon separates more than one reference by the same author full stop separates references by different authors figures of plates are enclosed in parentheses to distinguish them from text-figures dash, not comma, separates consecutive numbers Synonymy arrangement according to chronology of bibliographic references, whereby the year is placed in front of each entry, and the synonym repeated in full for each entry, is not acceptable. In describing new species, One specimen must be designated as the holotype; other speci- mens mentioned in the original description are to be designated paratypes; additional material not regarded as paratypes should be listed separately. The complete data (registration number, depository, description of specimen, locality, collector, date) of the holotype and paratypes must be recorded, e.g.: Holotype SAM-—A13535 in the South African Museum, Cape Town. Adult female from mid- tide region, King’s Beach Port Elizabeth (33°51’S 25°39’E), collected by A. ‘Smith, 15 January 1973. Note standard form of writing South African Museum registration numbers and date. 7. SPECIAL HOUSE RULES Capital initial letters (a) The Figures, Maps and Tables of the paper when referred to in the text e.g. ‘... the Figure depicting C. namacolus ...’; *. . . in C. namacolus (Fig. 10)...’ - (b) The prefixes of prefixed surnames in all languages, when used in the text, if not preceded by initials or full names e.g. Du Toit but A.L.du Toit; Von Huene but F. von Huene (c) Scientific names, but not their vernacular derivatives e.g. Therocephalia, but therocephalian Punctuation should be loose, omitting all not strictly necessary Reference to the author should be expressed in the third person Roman numerals should be converted to arabic, except when forming part of the title of a book or article, such as ‘Revision of the Crustacea. Part VIII. The Amphipoda.’ Specific name must not stand alone, but be preceded by the generic name or its abbreviation to initial capital letter, provided the same generic name is used consecutively. Name of new genus or species is not to be included in the title: it should be included in the abstract, counter to Recommendation 23 of the Code, to meet the requirements of Biological Abstracts. F. E. GRINE DESCRIPTION OF SOME JUVENILE HOMINID SPECIMENS FROM SWARTKRANS, TRANSVAAL RT 3 OCTOBER 1981 ISSN 0303-2515 © ANNALS OF THE SOUTH AFRIC CAPE TOWN. INSTRUCTIONS TO AUTHORS 1. MATERIAL should be original and not published elsewhere, in whole or in part. 2. LAYOUT should be as follows: (a) Centred masthead to consist of Title: informative but concise, without abbreviations and not including the names of new genera or species Author’s(s’) name(s) Address(es) of author(s) (institution where work was carried out) Number of illustrations (figures, enumerated maps and tables, in this order) (b) Abstract of not more than 200 words, intelligible to the reader without reference to the text (c) Table of contents giving hierarchy of headings and subheadings (d) Introduction (e) Subject-matter of the paper, divided into sections to correspond with those given in table of contents (f) Summary, if paper is lengthy (g) Acknowledgements (h) References (i) Abbreviations, where these are numerous 3. MANUSCRIPT, to be submitted in triplicate, should be typewritten and neat, double spaced with 2,5 cm margins all round. First lines of paragraphs should be indented. Tables and a list of legends for illustrations should be typed separately, their positions indicated in the text. All pages should be numbered consecutively. Major headings of the paper are centred capitals; first subheadings are shouldered small capitals; second subheadings are shouldered italics; third subheadings are indented, shouldered italics. Further subdivisions should be avoided, as also enumeration (never roman numerals) of headings and abbreviations. Footnotes should be avoided unless they are short and essential. Only generic and specific names should be underlined to indicate italics; all other marking up should be left to editor and publisher. 4. ILLUSTRATIONS should be reducible to a size not exceeding 12 « 18 cm (19 cm including legend); the reduction or enlargement required should be indicated; originals larger than 35 x 47 cm should not be submitted; photographs should be rectangular in shape and final size. A metric scale should appear with all illustrations, otherwise magnification or reduction should be given in the legend; if the latter, then the final reduction or enlargement should be taken into consideration. All illustrations, whether line drawings or photographs, should be termed figures (plates are not printed; half-tones will appear in their proper place in the text) and numbered in a single series. Items of composite figures should be designated by capital letters; lettering of figures is not set in type and should be in lower-case letters. The number of the figure should be lightly marked in pencil on the back of each illustration. 5. REFERENCES cited in text and synonymies should all be included in the list at the end of the paper, using the Harvard System (ibid., idem, loc. cit., op. cit. are not acceptable): (a) Author’s name and year of publication given in text, e.g.: ‘Smith (1969) describes...’ ‘Smith (1969: 36, fig. 16) describes .. .’ ‘As described (Smith 1969a, 19695; Jones 1971)’ ‘As described (Haughton & Broom 1927)...’ ‘As described (Haughton et al. 1927)...’ Note: no comma separating name and year Dagination indicated by colon, not p. names of joint authors connected by ampersand et al. in text for more than two joint authors, but names of all authors given in list of references. (b) Full references at the end of the paper, arranged alphabetically by names, chronologically within each name, with suffixes a, b, etc. to the year for more than one paper by the same author in that year, e.g. Smith (1969a, 19695) and not Smith (1969, 1969a). For books give title in italics, edition, volume number, place of publication, publisher. For journal article give title of article, title of journal in italics (abbreviated according to the World list o, scientific periodicals. 4th ed. London: Butterworths, 1963), series in parentheses, volume number, part number (only if independently paged) in parentheses, pagination (first and last pages of article). Examples (note capitalization and punctuation) BULLOUGH, W. S. 1960. Practical invertebrate anatomy. 2nd ed. London: Macmillan. FISCHER, P. —-H. 1948. Données sur la résistance et de le vitalité des mollusques. J. Conch., Paris 88: 100-140. FIsCHER, P.-H., DuvAL, M. & Rarry, A. 1933. Etudes sur les échanges respiratoires des littorines. Archs Zool. exp. Zen. 74: 627-634. Koun, A. J. 1960a. Ecological notes on @anus (Mollusca: Gastropoda) in the Trincomalee region of Ceylon. Ann. Mag. nat. Hist. (13) 2: 309-320. Konn, A. J. 19606. Spawning behaviour, egg masses and larval development in Conus from the Indian Ocean. Bull, Bingham oceanogr. Coll. 17 (4): 1-51. THIELE, J. 1910. Mollusca: B. Polyplacophora, Gastropoda marina, Bivalvia. In: SCHULTZE, L. Zoologische und anthropologische Ergebnisse einer Forschungsreise im westlichen und zentralen Siid-Afrika 4: 269-270. Jena: Fischer. Denkschr. med.-naturw. Ges. Jena 16: 269-270. (continued inside back cover) ANNALS OF THE SOUTH AFRICAN MUSEUM ANNALE VAN DIE SUID-AFRIKAANSE MUSEUM Volume 86 Band October 1981 Oktober Part 3 Deel MOSS S29 FS B:B.B.9.9) x > SS Youig nov NOVI DINOCEPHALIA TYPE MATERIAL IN THE SOUTH AFRICAN MUSEUM (REPTILIA, THERAPSIDA) By J. A. VAN DEN HEEVER KX Po Ee GRINE Cape Town Kaapstad The ANNALS OF THE SOUTH AFRICAN MUSEUM are issued in parts at irregular intervals as material becomes available Obtainable from the South African Museum, P.O. Box 61, Cape Town 8000 Die ANNALE VAN DIE SUID-AFRIKAANSE MUSEUM word uitgegee in dele op ongereelde tye na gelang van die beskikbaarheid van stof Verkrygbaar van die Suid-Afrikaanse Museum, Posbus 61, Kaapstad 8000 OUT: OF PRINT/UIT DRUK LOGS, 6), AGED, 755, 9, taal), SES, 5, 7-9), GG, taps), TA), 8, TER 1), 1003), GED, 57), jash)), ISS), DAO), Mh SIGS), WIG), 38, 5) EDITOR/REDAKTRISE Ione Rudner Copyright enquiries to the South African Museum Kopieregnavrae aan die Suid-Afrikaanse Museum ISBN 0 86813 020 6 Printed in South Africa by In Suid-Afrika gedruk deur The Rustica Press, Pty., Ltd., Die Rustica-pers, Edms., Bpk., Court Road, Wynberg, Cape Courtweg, Wynberg, Kaap DINOCEPHALIA TYPE MATERIAL IN THE SOUTH AFRICAN MUSEUM (REPTILIA, THERAPSIDA) By J.A. VAN DEN HEEVER* & F.E. GRINE South African Museum, Cape Town [MS accepted 14 May 1981] ABSTRACT Brief statements on the preservation of the dinocephalian type specimens, the localities from which they were recovered, and the collectors responsible are given. The original generic and/or specific diagnoses are reproduced. All relevant references to each specimen are listed, together with pertinent comments. CONTENTS PAGE EAT PRO CUETO let Pa ede eg eae hare chau tUnS, ekcty ofa eon Gene 73 Sy Meat atetial tee. secene sty ei Meer nat a tec earteR e esete A Suche 74 amilyeANteOSaUnidae, yesh see eek cel oe ee eis eke 74 Ramilyaiitanosuchidac.. eneai aia: scene acess see Gae oe 85 amily skapimocephalidaes 225 22 /ssciretea! sade eee AB 98 Subfamily, tapmocephalinaes 235-25 -ssse eee esate 3 98 Subfamuly;struthiocephalinaes5.0 44-ee aee te ae o- 99 SubfamuilysRiebeeckosauninaens ee eee ea ee 104 Subfamily Moschopinac sy 94-4ae oe ee ace: 106 RamulysStyracocephalidae 25.245 5. tlitede soe dee Gee oe he bit PACKNOWICUSCMIETIES cs ec eee. c citer ureaers eee ars Seis a hee eee 112 INCHCTOM COS oe SAG oh aime ete he item ete am ae ar N28 Oe a alee 112 INTRODUCTION The infra-order Dinocephalia (assuming that this group is monophyletic) represents a Middle Permian radiation of carnivorous and herbivorous ther- apsids. In the Tapinocephalus Zone (Kitching 1977) sediments of the Beaufort Group, Cape Province, South Africa, the dinocephalians form a substantial part of the dominant therapsid fauna. It is commonly held that the South African dinocephalians are derived from earlier Russian therapsids (Olson 1962; Boonstra 1969; Tatarinov 1974, 1976). Since the South African Museum houses the largest collection of dinoce- phalian type specimens, it was felt that a comprehensive list of this material would serve a useful purpose. The collection contains some forty-four holotype * The order of authorship was decided by the flip of a coin. WS Ann. S. Afr. Mus. 86 (3), 1981: 73-114. 74 ANNALS OF THE SOUTH AFRICAN MUSEUM and paratype specimens and, in order to present this list of material in a clear manner, Boonstra’s (1969) familial and subfamilial classifications have been utilized. The generic and specific diagnoses included in this paper are taken from the original diagnosis of each taxon. Those that have been indicated clearly as diagnoses are reproduced verbatim. In some instances, where the taxonomic diagnosis was not clearly separated from the descriptive text, the present authors have scrutinized the description in order to determine which features were considered by the original author to be diagnostic. However, in all cases the terminology remains that of the original author. Most of the localities are represented by farm names, spelt in a variety of -ways by different workers. In order to avoid confusion, the authors have followed Kitching (1977) in the spelling of locality names. TYPE MATERIAL Family Anteosauridae Anteosaurus abeli Boonstra, 1952 Holotype SAM-11296 Material Fairly complete skull and lower jaw which have been distorted slightly by shear. Locality Kruisrivier, Sutherland. Collected Boonstra and Laurenson, 1937. Generic diagnosis See Watson (1921) for diagnosis of Anteosaurus. Specific diagnosis Skull large and massive; postorbital boss less massive than in genotype [sic]; jugal boss low to very prominent; bone thickened around pineal foramen but does not form a clear, circular boss; frontals strongly swollen; tooth-bearing palatine boss semilunate; premaxillary edge slopes upward in contrast to genotype [sic]; incisors very long with only some showing slight indication of lingual step; premaxillae greatly swollen dorsally and clearly demarcated from maxillae; squamosal moderately flared laterally; dentary strong and very mass- ive (after Boonstra 1952e: 150). DINOCEPHALIA TYPE MATERIAL TS References Boonstra 1948: 40 (fig.). Boonstra 1952e: 150-151. Boonstra 1953a: 26, pl. 1. Boonstra 1954a: 109, 112-114, 124-125, 144, figs 1-2. Haughton & Brink 1955: 42. Piveteau 196la: 84, fig. 11. Comments Boonstra (1969) regarded Anteosaurus abeli as a junior synonym of Anteo- saurus magnificus. Paratype SAM-—340 Material Good skull and part of lower jaw which have been distorted by shear. Locality Leeurivier, Beaufort West. Collected Haughton, 1916. References Boonstra 1952e: 150. Boonstra 1953a: pl. 3. Boonstra 1954a: 108, 112-113, 124, 130, 144, fig. 6. Comments An isolated symphysis labelled SAM—4340 does not belong to this individ- ual. There are several unlabelled fragments associated with the specimen which do not appear to belong to it. Paratype SAM-5621 Material Incomplete, fragmentary skull consisting of part of snout and a portion of cranial roof. Locality Leeurivier, Beaufort West. 76 ANNALS OF THE SOUTH AFRICAN MUSEUM Collected Haughton, 1916. References Boonstra 1952e: 150. Boonstra 1954a: 108, 131, 144. Paratype SAM-9123 Material Weathered and sectioned skull lacking lower jaw; two vertebrae. Locality Voélfontein, Prince Albert. Collected Boonstra, 1929. References Boonstra 1952e: 150. Boonstra 1954a: 108, 131, 144. Boonstra 1962: 97-98; fig. 33. Paratype SAM-11302 Material Fairly complete but weathered skull with part of lower jaw. Locality Buffelsvlei, Beaufort West. Collected Boonstra, 1938. References Boonstra 1952e: 150. Boonstra 1954a: 134-135, 144. Comments Boonstra (1954a) placed this specimen in the hypodigm of Anteosaurus crassifrons. DINOCEPHALIA TYPE MATERIAL Paratype SAM-11929 Material Imperfect, weathered skull. Locality See comments. Collected See comments. References Boonstra 1952e: 150. Boonstra 1954a: 109, 135, 144. Comments ql The locality is questionable. In the museum catalogue it is given as ‘unknown, probably near Abrahamskraal, Prince Albert’. Boonstra (1954a: 109) gave the locality as Kruisvlei, Beaufort West, but later in the same paper he (1954a: 135) referred to the locality as Abrahamskraal, Prince Albert. The museum catalogue records the collector as ‘unknown’; Boonstra (1954a: 109) stated that he collected it. Boonstra (1954a) placed this specimen in the hypodigm of Anteosaurus crassifrons. Anteosaurus acutirostris Boonstra, 1954 Holotype SAM-9329 Material Nearly complete skull and lower jaw. Locality Kruisvlei, Beaufort West. Collected Boonstra, 1929. Generic diagnosis See Watson (1921) for diagnosis of Anteosaurus. Specific diagnosis Skull large, maximum length of 675 mm. Snout long, high, narrow, and light. Intersquamosal width great (480? mm). Postfrontal boss huge and pro- 78 ANNALS OF THE SOUTH AFRICAN MUSEUM minent. Jugal boss massive, angular boss massive. Fronto-naso-prefrontal swell- ing very strong with distinct step on to the anterior nasal surface and laterally slightly overhanging the sides of the skull. Pineal boss low with rounded edges extending to the occipital edge. Fairly strong upward inclination of premaxillary edge. Occiput high, fairly wide, very deeply concave with sharp and great posterior sweep of temporal arches; upper part of temporal fossa roomy anteroposteriorly. Temporal arch rises above plane of intertemporal surface. Palate apparently long and narrow. Basis cranu fairly long. Exocipitals fused with basioccipital to form rounded condyle, which is visible beyond occipital edge in dorsal view. On both sides 4 upper and 4 lower incisors, 1 canine, 5? postcanines. (After Boonstra 1954a: 131-132.) References Boonstra 1952e: 150. Boonstra 1953a: pl. 2. Boonstra 1954a: 108, 113, 131-133, 144, fig. 10. Haughton & Brink 1955: 42. Comments Boonstra (1969) regarded Anteosaurus acutirostris as a junior synonym of Anteosaurus magnificus. Anteosaurus crassifrons Boonstra, 1954 Holotype SAM-11946 Material Good skull distorted by shear. Locality Buffelsvlei, Beaufort West. Collected Boonstra and Marais, 1951. Generic diagnosis See Watson (1921) for diagnosis of Anteosaurus. Specific diagnosis Skull large but short and squat; maximum length 570 mm. Snout short, high and very wide. Intersquamosal width fairly small (330? mm). Postfrontal boss fairly massive and prominent. Jugal boss massive. Fronto-naso-prefrontal swelling very massive with very distinct step on to anterior nasal surface and DINOCEPHALIA TYPE MATERIAL 79 laterally strongly overhanging sides of skull. Pineal boss rounded, large and extending on to frontal. Very sharp inclination of premaxillary edge. Occiput high, fairly wide, deeply concave, with great posterior sweep of temporal arches. Upper part of temporal fossa roomy in anteroposterior direction. Temporal arch not rising above plane of intertemporal surface. Palate very short with massive lateral pterygoid flanges. Basis cranii short, with short basisphenoid. Exoccipitals fused with basi-occipital to form rounded condyle. In both premaxillae teeth not preserved but five matrix-filled alveoli shown; in both maxillae five postcanine roots of greatly varying diameter preserved, numbers 2 and 5 much smaller than other three. (After Boonstra 1954a: 133-134.) References Boonstra 1954a: 109, 113, 133-134, 144, figs 11-13. Haughton & Brink 1955: 42. Comments Boonstra (1969) regarded Anteosaurus crassifrons as a junior synonym of Anteosaurus magnificus. Anteosaurus cruentus Boonstra, 1954 Holotype SAM-11694 Material Reasonably complete skull without lower jaw. Locality Koringplaas, Laingsburg. Collected Boonstra and Du Plessis, 1946. Generic diagnosis See Watson (1921) for diagnosis of Anteosaurus. Specific Diagnosis Skull moderately large, maximum length 565 mm. Snout long, narrow and fairly low. Intersquamosal width not great (360 mm). Postfrontal boss fairly strong and prominent. No jugal boss. Fronto-naso-prefrontal swelling small, passing evenly on to anterior nasal surface. Pineal boss prominent, with sharp circular border, situated very near occipital edge. Sharp upward inclination of premaxillary edge. Occiput high and fairly broad, very deeply concave with 80 ANNALS OF THE SOUTH AFRICAN MUSEUM great posterior sweep of temporal arch. Upper part of temporal fossa roomy anteroposteriorly; temporal arch not rising above very narrow intertemporal surface. Palate long, with fairly robust lateral pterygoidal flanges. Basis cranii long, but basisphenoid short. Exoccipitals form large part of dorsolateral corners of condyle. Alveolar face of premaxilla shows matrix-filled groove divided in its posterior part by distinct alveoli; appears to have been room for 5 incisors when fully developed. On right side canine root followed by roots of 6 postcanines, but on left only 4 roots with a possible fifth visible. (After Boonstra 1954a: 139-140.) References Boonstra 1953a: 26, pl. 6. Boonstra 1954a: 109, 112, 114, 124, 139-141, 144, figs 17-20. Haughton & Brink 1955: 42. Comments Boonstra (1953a) referred this speciman to A. minor. Boonstra (1969) regarded Anteosaurus cruentus as a junior synonym of A. magnificus. Anteosaurus levops Boonstra, 1954 Holotype SAM-11492 Material Badly weathered, incomplete skull without lower jaw. Locality Mynhardtskraal, Beaufort West. Collected Boonstra and Bothma, 1940. Generic diagnosis See Watson (1921) for diagnosis of Anteosaurus. Specific diagnosis Skull fairly small, maximum length 485 mm. Snout fairly short, lightly built, narrow and low. Intersquamosal width relatively large (415 mm). Post- frontal boss strong and prominent. No jugal boss. Fronto-naso-prefrontal swelling strong, with distinct step on to anterior nasal surface and laterally slightly overhanging sides of skull. Pineal boss apparently prominent, reaching occipital edge. Upward inclination of premaxillary edge moderate. Occipital fairly low and broad, deeply concave, not vertical; strong posterolateral sweep DINOCEPHALIA TYPE MATERIAL 81 of temporal arches. Upper part of temporal fossa shortened in anteroposterior direction; temporal arch rising above plane of narrow intertemporal surface. Palate long and narrow. Basis cranii fairly short. In right premaxilla parts of crowns of 3 incisors preserved, but on left there is a matrix-filled groove with no sign of teeth. (After Boonstra 1954a: 141-142.) References Boonstra 1953a: 26. Boonstra 1954a: 109, 141-142, 144, fig. 21. Haughton & Brink 1955: 43. Comments Boonstra (1953a) considered this skull to be a referred specimen of A. minor. Boonstra (1969) regarded A. Levops as a junior synonym of A. magnificus. Anteosaurus laticeps Boonstra 1954 Holotype SAM-11592 Material Incomplete skull with only ventral and occipital surfaces preserved. Locality Dikbome, Laingsburg. Collected Boonstra and Du Plessis, 1942. Generic diagnosis See Watson (1921) for diagnosis of Anteosaurus. Specific diagnosis Skull large, maximum length 645? mm. Snout short and very broad. Intersquamosal width great (522 mm). Occiput low and very broad, shallowly concave; sweep of temporal arches mostly lateral. Palate long and very broad with only moderately strong lateral pterygoidal flanges. Basis cranii short, with very short basisphenoid. Exoccipitals form much of dorsolateral corners of condyle. Canines very strong and slightly recurved. In right maxilla, stumps of 5 postcanines can be made out whereas on left 5 teeth are preserved with the probability of another 2, making a total of 7. Postcanines bluntly conical, but linguolabially compressed with crowns thus oval in cross-section (16 x 11 x 7 mm). (After Boonstra 1954a: 138-139.) 82 ANNALS OF THE SOUTH AFRICAN MUSEUM References Boonstra 1952e: 150. Boonstra 1953a: 22. Boonstra 1954a: 109, 138-139, 144, fig. 16. Haughton & Brink 1955: 42. Comments Boonstra (1952e) made this specimen a paratype of A. abeli. Later he (1953a) appears to infer that he considers it to be a specimen of Jonkeria sp. Boonstra (1969) regarded A. laticeps as a junior synonym of A. magnificus. Anteosaurus major Boonstra, 1954 Holotype SAM-11293 Material Good skull, dorsoventrally compressed, lacking snout. Locality Boesmansrivier, Beaufort West. Collected Boonstra, 1937. Generic diagnosis See Watson (1921) for diagnosis of Anteosaurus. Specific diagnosis Skull very large, maximum length 805? mm. Snout long, fairly broad and high. Intersquamosal width very great (612 mm). Postfrontal boss only mod- erately strong and not very prominent, with postorbital forming posterolateral part. Low and weak jugal boss. Fronto-naso-prefrontal boss moderate, con- fluent with anterior nasal surface. Pineal boss low, with rounded edges, situated some distance from occipital edge. Occiput low and broad, moderately deeply concave. Sweep of temporal arches more lateral than posterior and upper part of temporal fossa roomy anteroposteriorly. Temporal arch rises above level of intertemporal surface. Palate fairly long and broad, with strong and deep lateral pterygoidal flanges but not so massive as in A. crassifrons. Basis cranii- long and basisphenoid much longer than in any of the other species. Exoccipi- tals from dorsolateral corners of condyle. (After Boonstra 1954a: 136-138.) References Boonstra 1952e: 150. Boonstra 1953a: pls 4-5. DINOCEPHALIA TYPE MATERIAL 83 Boonstra 1954a: 109, 120, 136-138, 144, figs 14-15. Haughton & Brink 1955: 42. Comments Boonstra (1952e) referred to this specimen as a paratype of A. abeli. Boonstra (1969) synonymized this taxon with Anteosaurus magnificus. Micranteosaurus parvus Boonstra, 1954 Holotype SAM-4323 Material Anterior part of snout, lower jaw, manus, pes, femur, fibula, proximal part of humerus, radius, coracoid, caudal vertebra. Locality Merweville commonage, Beaufort West. Collected Haughton, 1917. Generic diagnosis Anteosaurid of very small size (after Boonstra 1954b: 156). Specific diagnosis As for genus. References Boonstra 1954b: 149-156, figs 1-4, pl. 18. Boonstra 19555: 200, 320-321. Haughton & Brink 1955: 45. Von Huene 1956: 287. Piveteau 196la: 85. Orlov 1964: 254. Boonstra 1966: 14, 17-18, 23-25, figs 3, 10. Boonstra 1969: 33. Comments Boonstra (1969) regarded this specimen as a juvenile Anteosaurus magnificus. Paranteosaurus primus Boonstra, 1954 Holotype SAM-11485 84 ANNALS OF THE SOUTH AFRICAN MUSEUM Material Partial, fragmented skull, proximal end of femur, vertebrae. Locality Mynhardtskraal, Beaufort West. Collected Boonstra, 1940. Generic diagnosis ‘A medium-sized Anteosaurid [sic] (max. length probably about 570 mm), with small postfrontal not extending posteriorly, lightly built postorbital bar, without any sign of a boss-like development in the upper part of the postorbital bar, dental formula I.3?-4?, C.1, P.C.5’ (Boonstra 1954c: 159). Specific diagnonis As for genus. References Boonstra 1952e: 150. Boonstra 1954c: 157-159. Boonstra 19555: 320, fig. 102e. Haughton & Brink 1955: 45. Piveteau 1961a: 85. Boonstra 1963a: 177, fig. 8G. Boonstra 1963b: 200, fig. 4A. Orlov 1964: 254. Boonstra 1969: 33, 35, 55. Eccasaurus priscus Broom, 1909 Holotype SAM-915 Material Humerus, two imperfect femora, fibula, distal end of tibia, occipital fragment, tooth, nine weathered vertebrae, rib fragments, nine additional badly preserved fragments. Locality Sandvlakte, Prince Albert. Collected Presented by Cairncross. DINOCEPHALIA TYPE MATERIAL 85 Generic diagnosis Humerus with well-developed deltopectoral ridge which does not continue to head (after Broom 1909a: 276). Specific diagnosis As for genus. References Broom 1909a: 276-277. Gregory 1926: 235. Broom 1932: 45, fig. 3c-d. Boonstra 19556: 188, 321-322, figs 105-106, 107a. Haughton & Brink 1955: 47. Romer 1966: 373. Boonstra 1969: 33. Comments Broom (1909a) considered the humerus to be most similar to that of Procolophon, but felt Eccasaurus to be a primitive diaptosaurian which ‘prob- ably belongs to a new suborder’. In 1932 he expressed the opinion that Eccasaurus is undoubtedly a tapinocephaloid and noted that its humerus resembles, in both morphology and size, that of Moschops. Boonstra (1955b) considered Eccasaurus to be an anteosaurid, while Romer (1966) placed the genus in the family Tapinocephalidae. Boonstra (1969) regarded the taxon to be determinable only to family. Family Titanosuchidae Archaeosuchus cairncrossi Broom, 1905 Holotype SAM-916 Material Maxilla, weathered and incomplete (side questionable). Locality Sandvlakte, Prince Albert. Collected Cairncross. Generic diagnosis Broom (1905) did not give a diagnosis but commented that the canine looked titanosuchian. Broom (1932: 18) stated that: ‘The canine is remarkable 86 ANNALS OF THE SOUTH AFRICAN MUSEUM in having a posterior ledge. The molars are relatively small, short conical teeth.’ Specific diagnosis As for genus. References Broom 1905: 333-335. Broom 1909b: 287. Broom 1932: 18, fig. 3J. Boonstra 1953a: 28. Haughton & Brink 1955: 46. Comments Broom (1905) provisionally placed this taxon in the Therocephalia, but in 1909 he put it in the Dinocephalia. Broom (1932) commented that it was perhaps representative of a ‘pre-dinocephalian group’. Boonstra (1953a) trans- ferred this specimen to Titanosuchus and questioned its specific validity. Boon- stra (1969) noted that the specimen was so poor that it must be regarded as a nomen dubium. Dinosphageus haughtoni Broom, 1929 Holotype SAM-—4343 Material Humeri, ilia, good skull without lower jaw, scapula, cleithrum, ulna, ischium, femur, ?fibula, coracoids and vertebrae. Locality Welgemoed, Leeu-Gamka, Prince Albert. Collected Haughton, 1916. Generic diagnosis Vomers form marked median ridge on their posterior two-thirds. Ptery- goids project much less downward than in Jonkeria. Upper half of humerus little more in plane of lower half than is the case in humerus of Jonkeria and, as a result, on side view the deltopectoral crest appears smaller, but this may be due to crushing. Upper portion of ilium relatively smaller than in Jonkeria. (After Broom 1929: 31.) Specific diagnosis As for genus. DINOCEPHALIA TYPE MATERIAL 87 References Broom 1929: 31, figs 17-19. Broom 1932: 29, fig. 7D. Boonstra 1953a: 27. Boonstra 19555: 189, 287-293, figs 75-80. Haughton & Brink 1955: 44. Boonstra 1962: 78-80, fig. 18. Boonstra 19655: 263, fig. 9. Boonstra 1969: 37. Comments Broom (1932) emended the diagnosis. Boonstra (1953a) transferred the species to Jonkeria. No mention is made in the literature of vertebrae belonging to the specimen but several vertebrae were found bearing the number SAM-4343 and probably belong to the specimen. Jonkeria koupensis Boonstra, 1955 Holotype SAM-9004 Material Pelvis lacking ischia; distal two-thirds of ulna. Locality Klein Koedoeskop, Beaufort West. Collected Boonstra, 1929 Generic diagnosis See Van Hoepen (1916) for diagnosis of Jonkeria. Specific diagnosis Pubo-ischiadic plate probably short (83 per cent of height as recon- structed); supra-acetabular part of ilium high (264 mm) and anteroposterior length of iliac blade very short (282 mm), so that height is 93 per cent of length; anterior iliac process fairly short, but appears shorter than it really is because of strong anterolateral eversion and it is fairly high; posterior process short and fairly low, with posteroventral edge moderately strongly folded over to form a fairly strong iliofibularis ridge on outer face; this ridge is directed obliquely upward, with its upper end not strongly bulbously thickened; a slight groove on the inner face of the everted anterior iliac process indicates attachment of rib anterior to main sacral rib; anteroposteriorly the outer face of iliac blade deeply 88 ANNALS OF THE SOUTH AFRICAN MUSEUM concave. Anteroventral edge of pubis strongly everted, with tuberculum pubis confluent with thickened part of anteroventral edge, which stretches to median line where the pubes meet but do not form a real symphysis. Ulna massive with dorsal lip of sigmoid face developed into massive swelling in its pre-axial part. (After Boonstra 1955b: 301.) References Boonstra 19556: 301, fig. 87. Boonstra 1969: 38. Jonkeria parva Boonstra, 1955 Holotype SAM-9149 Material Well-preserved right humerus. Locality Saairivier, Prince Albert. Collected Boonstra, 1929. Generic diagnosis See Van Hoepen (1916) for diagnosis of Jonkeria. Specific diagnosis Smallest Jonkeria humerus known; length only 312 mm, but it is a very massive element with both proximal (222 mm) and distal (252 mm) ends very greatly expanded; shaft very short and bone greatly constricted in waist; diameters of shaft 84 x 78 mm; deltopectoral crest fairly short and terminates very far proximal of the plane in which entepicondylar foramen lies; it has a very massive ventral edge, and terminates as a very thick knob. Caput weak and its face strap-like, but it forms the most proximal part of the bone; processus lateralis lies more distal than the caput; processus medialis, as in all Jonkerias [sic], lies well distal of caput; capitellum very massive and extends very far proximally along ventral face, with its proximal border lying in a plane proximal to that in which the entepicondylar foramen lies; posterior to capitellum there is a deep groove in which the coronoid process moved when the ulna was flexed; twist of shaft great (30°); lateromedial line distinct with large, mound-like muscle scar on dorsal surface of shaft; anterior dorsoventral line very strong and forms prominent ridge; entepicondyle strongly developed to form a greatly outflaring, thick sheet of bone; ventral opening of entepicon- DINOCEPHALIA TYPE MATERIAL 89 dylar foramen oval and lies well postaxially, near edge of bone; ectepicondyle developed as a greatly flaring thin sheet of bone penetrated in its thinner part near edge by small, round ectepicondylar foramen. (After Boonstra 1955p: 303.) References Boonstra 1955b: 192, 303, fig. 89. Boonstra 1969: 38. Jonkeria rossouwi Boonstra, 1955 Holotype SAM-5014 Material Left scapula, incomplete precoracoid and coracoid, imperfect inter- clavicle, left humerus, left ulna, left radius, right ilium, left femur, left tibia, left fibula. Locality Abrahamskraal, Prince Albert. Collected Haughton per Van der Byl. Generic diagnosis See Van Hoepen (1916) for diagnosis of Jonkeria. Specific diagnosis Pectoral girdle fairly large and fairly massive; scapula fairly low (height 552 mm) and upper part of blade greatly expanded (width 324 mm), tricipital bulge very prominent, supraglenoidal edge forms strong raised rim; internal opening of supracoracoid foramen opens into deep subscapular groove; glenoidal facet of scapula faces ventroposteriorly but not externally; precoracoid long but low, foramen pierces bone very obliquely. Coracoid small but massive, with large glenoidal facet facing well externally. Interclavicle massive but short (480? mm) with stem wide posteriorly and with a narrowed waist anteriorly; anterior spatulate end curves upwards very sharply and has deep groove on outer anterolateral face for reception of posteroventral edge of clavicle; on dorsal surface of stem these is a strong medial ridge against which the precoracoids abut. Humerus fairly short (378 mm in length), but massive with greatly expan- ded proximal (330 mm) and distal (276? mm) ends; shaft very short, thick and broad (132 x 84 mm); deltopectoral crest very long and nearly reaches plane in 90 ANNALS OF THE SOUTH AFRICAN MUSEUM which entepicondylar foramen lies; caput broadly oval; processus lateralis lies well proximally, in same plane as caput, whereas processus medialis lies well distally; capitellum fairly massive and extends well along ventral face and nearly reaches plane of entepicondylar foramen; ‘twist’ on shaft fairly small (15°); lateromedial line fairly strong with muscle scars on dorsal surface of shaft; anterior dorsoventral line well developed; entepicondyle strong with ventral opening of foramen large and nearly round; ectepicondyle forms thick flange pierced by small foramen situated well away from edge of bone. Ulna pathological but the normal condition would appear to have had a more slender shaft and weaker coronoid process than in other known species of Jonkeria. Radius (length 312 mm) with flange on proximopostaxial corner weak. Supra-acetabular part of ilium high (288 mm) and relatively short (336 mm) so that height is 87 per cent of length; anterior iliac process relatively short, but fairly high and strongly everted; posterior process short and fairly low, with posteroventral edge folded over strongly to form strong vertical ridge, which is dorsally strongly bulbous, and projects strongly laterally; on inner face of anterior iliac process no distinct facet is preserved for attachment of rib lying anterior to main sacral rib. Femur fairly long (504 mm); fairly broad over external trochanter (264 mm), which is not separated by notch from proximal face; caput fairly thick (114 mm) and directed much preaxially; shaft broad but flat (156 x 84 mm); femorotibialis ridge fairly strong; distal facets of femur small and directed much distally, especially the entepicondyle which lies far distally. Tibia fairly robust (length 300 mm); proximal face inclined much postax- ially to correspond with the distally situated postaxial facet of femur. Fibula fairly slender and long (330 mm). (All after Boonstra 1955b: 303-305, 309.) References Broom 1929: 27-29, figs 15-16. Boonstra 1955b: 303-305, 309, figs 90-97. Comments Broom (1929) referred to this specimen as Jonkeria sp. and merely noted the presence of other elements. Parascapanodon avifontis Boonstra, 1955 Holotype SAM-9127 Material Good precoracoid, clavicle, fibula and femur associated with parts of a large skull. DINOCEPHALIA TYPE MATERIAL 91 Locality Voélfontein, Prince Albert. Collected Boonstra, 1929. Generic diagnosis ‘The generic characters of the pectoral girdle are as described for the family [Titanosuchidae]’ (Boonstra 1955b: 274), i.e. pectoral girdle large and massive .. . length of coracoidal plate 62-67 per cent of height. Scapula high and upper end of blade broad to very broad; scapular head of triceps attached to sharp ridge or prominent mound. Precoracoid large and massive, coracoid of medium size but massive. Clavicle large and mediolaterally flattened with expanded dorsal and ventral ends but waist not greatly constricted, ventral spatulate end curves inward to fit over outer face of upturned anterolateral corner of interclavicle; dorsal anterior end greatly thickened and produced dorsally as a short, strong process which fits into groove on lower end of cleithrum. Interclavicle large and massive, lateral horns with thickened postero- lateral edges, anterior to which there is a fairly deep groove to house ventro- posterior edge of spatulate end of clavicle. Humerus very large and massive, length 575 mm and ends greatly expan- ded (proximal 310? mm, distal 312 mm); shaft fairly long but very robust (144 x 142 mm); deltopectoral crest long but terminates well proximal of plane in which entepicondylar foramen lies; caput very massive but short; processus medialis lies just a little distally of a plane in which caput lies, capitellum very strong and massive and extends far along ventral face but does not reach plane of entepicondylar foramen; twist on shaft large (40°); lateromedial line strong with massive swelling on dorsal surface of shaft, both epicondyles strongly developed; ventral opening of entepicondylar foramen large and broadly oval, ectepicondylar foramen small and situated well away from edge of bone. Ulna large and massive (length 372 mm, width over coronoid process 200 mm); sigmoid face long, with ventral part broadly rounded; coronoid process situated far distally and shaft massive, broad and short. Radius long, robust (length 294 mm) with strong proximopostaxial flange. Femur very long and massive (length 595 mm); very broad over external trochanter (300 mm); pre-axial face deeply concave, with caput much pre- axially directed and massive (diams 215 x 167 mm); external trochanter indis- tinctly separated by notch from proximal face; shaft fairly long and broad (breadth 150 mm); wide over massive distal facets; area of origin of femoroti- bialis forms strong bulging ridge. Tibia large and massive (length 330-355 mm); cnemial eminence very massive and continued distally as strong ridge, with deep groove lying post- axially. 92 ANNALS OF THE SOUTH AFRICAN MUSEUM Fibula large and stout (length 330-345 mm). (All after Boonstra 1955b: 266-267, 274, 276.) Specific diagnosis As for genus. References Boonstra 1955b: 266-267, 274, 276, 278, figs 63-66. Boonstra 1962: 69, figs 10, 39A. Comments The skull is not described, illustrated or considered in the diagnosis. Boonstra (1969) considered all characters described under the names Titanosu- chus, Scapanodon, and Parascapanodon as diagnostic of the form Titanosuchus ferox. Paratype SAM-9106 Material Incomplete scapula, coracoid and precoracoid. Locality Veldmansrivier, Prince Albert. Collected Boonstra, 1929. References Boonstra 1955b: 278, fig. 67. Paratype SAM-9163 Material Good fibula and fair ulna. Locality Wakkerstroom (part of Wolwefontein), Prince Albert. Collected Boonstra, 1929. References Boonstra 1955b: 280, fig. 70a-e. DINOCEPHALIA TYPE MATERIAL 93 Paratype SAM-11299 Material Right tibia and right radius. Locality Boesmansrivier, Beaufort West. Collected Boonstra, 1938. References Boonstra 1955b: 281, fig. 71a-f. Paratype SAM-11488 Material Interclavicle, precoracoids, coracoid, right tibia, right fibula. Locality Mynhardtskraal, Beaufort West. Collected Boonstra, 1940. References Boonstra 19555: 278, figs 68, 69a-f. Comments According to the museum catalogue, the locality is Mynhardtskraal, Beau- fort West. Boonstra (1955) recorded the locality as Voélfontein, Prince Albert. On both the tibia and fibula the number SAM-—9123 has been scratched out. SAM-9123 is the paratype of A. abeli, a skull from Voélfontein, Prince Albert. It would appear, therefore, that Mynhardtskraal is the correct locality. Paratype SAM-11881 Material Humerus. Locality Bloukrans, Prince Albert. 94 ANNALS OF THE SOUTH AFRICAN MUSEUM Collected Boonstra, 1948. References Boonstra 1955b: 284, fig. 72. Scapanodon duplessisi Broom, 1904 Holotype SAM-—769 Material Three lower jaw fragments. Locality Seekoeigat, Prince Albert. Collected Presented by Du Plessis. Generic diagnosis Two fairly large incisors, very large canine and (at least) eleven molars. Molars considerably smaller and much flatter than in Titanosuchus. Molars two-thirds of size of Titanosuchus. Section of deeper part of molar root narrow oval and becomes flatter as approaches alveolar margin. Supra-alveolar part of tooth has fair-sized root apparently devoid of enamel, with edges moderately parallel. In crown the flattening continued to even greater extent. Thickest part of crown not more than 2,0 mm thick and from centre it thins off towards edges. Tooth strengthened by being slightly concavoconvex. External surface not grooved though slightly uneven, no serrations visible at edges. Enamel very thin (about 0,15 mm). (After Broom 1904: 182-183.) Specific diagnosis As for genus. References Broom 1904: 182-183. Broom 1923: 663. Broom 1932: 30. Boonstra 1953a: 25, 28. Haughton & Brink 1955: 46. Boonstra 1969: 35. Comments Broom (1904) described two of the fragments, which belong to the same lower jaw. The other piece, which has a single root and which may belong to DINOCEPHALIA TYPE MATERIAL 05 the same individual, was not described. Boonstra (1953a: 25) transferred the specimen to Titanosuchus but on page 28 regarded it as incertae sedis. Boonstra (1955a) resurrected Scapanodon on the basis of S. septemfontis. Haughton & Brink (1955: 46) retained SAM-—769 in Titanosuchus. Boonstra (1969: 35) again regarded it as Titanosuchus, but considered the cranial features of this taxon to be indeterminate. He stated (1969: 35) that the humerus could not be distin- guished from that of Titanosuchus, whilst the skull referred to this genus by Broom (1923) falls within the limits set by Boonstra for the genus Jonkeria. Scapanodon septemfontis Boonstra, 1955 Holotype SAM-S001 Material Ilium, pubis, femur, humerus, ischium. Locality Sewefontein, Prince Albert. Collected Haughton, 1917. Generic diagnosis See Broom (1904) for diagnosis of Scapanodon. Specific diagnosis Humerus large, fairly long (480? mm); deltopectoral crest terminates far distally but still well away from ventral opening of entepicondylar foramen (after Boonstra 195556: 271-272). Pelvis and femur as in generic diagnosis, as emended by Boonstra (19555: 269-270). (See comments.) References Broom 1928: 431, figs 4, 5A. Boonstra 19556: 271-272, figs 60a, 61-62. Comments Broom (1904: 182) mentioned several other bones, including two humeri (SAM-772 and SAM-773) found in the same area as the type of S. duplessisi (SAM-769). Boonstra (19555) considered these specimens to belong to the same species as SAM-769 and based this new species (S. septemfontis) on the different humeral structure of SAM-5001. Boonstra (19555) emended the generic diag- nosis to include postcranial elements. Broom (1928) referred SAM-—5001 to 96 ANNALS OF THE SOUTH AFRICAN MUSEUM Tapinocephalus atherstoni, while Boonstra (1955b) made it the type of a new species of Scapanodon, but earlier he (1953a) regarded Scapanodon (as known from Broom’s S. duplessisi) as a junior synonym of Titanosuchus. Boonstra (1969), however, regarded Scapanodon as a junior synonym of Titanosuchus ferox. Titanosuchus cloetei Broom, 1903 Holotype SAM-731. Material Left anterior part of lower jaw. Locality Gamka River, Prince Albert. Collected Presented by Cloete. Generic diagnosis See Owen (1879) for diagnosis of Titanosuchus. Specific diagnosis Differentiated from 7. ferox principally by dental measurements. Incisors in 7. cloetei appreciably smaller, canine much smaller and rounder. In T. cloetei 4 molars occupy space of 40 mm, each root practically round and about 8,0 mm in diameter. Another difference is arrangement of anterior molars. In T. ferox the line of the molars is much on the inside of the large canine, but in T. cloetei a line drawn along inner sides of molars also forms a tangent to the canine. Front of jaw also very much squarer in smaller species owing to canine being relatively further forward. (After Broom 1903: 142-143.) References Broom 1903: 142-143. Broom 1909b: 287. Haughton 1915): 57. Broom 19293 11 25.aien 25G- Broom 1932: 24, fig 6G. Boonstra 1953a: 27. Haughton & Brink 1955: 47. Boonstra 1962: 76-77, fig. 16. Boonstra 1969: 35. Kitching 1977: 36. DINOCEPHALIA TYPE MATERIAL 97 Comments Broom (1929) transferred the specimen to Jonkeria. Boonstra (1953a) retained it as Titanosuchus cloetei. Boonstra (1969) noted that the specimen was determinable only to the family Titanosuchidae. Titanosuchus dubius Haughton, 1915 Holotype SAM-2759 Material Right ramus of lower jaw. Locality Platfontein, Prince Albert. Collected Haughton, 1913. Generic diagnosis See Owen (1879) for diagnosis of Titanosuchus. Specific diagnosis Differs from T. cloetei in having a more massive symphysis, in being even squarer in the front of the jaw, and in the much larger canine and much smaller molars. Incisors similar in size but set much closer together. Canine narrower and longer, almost oblong in section with one side more than twice the length of the other. Molars smaller and circular in section. Line drawn along the inner side of molars just touches canine and is also tangential to inner surface of the fourth incisor. (After Haughton 19155: 57.) References Haughton 19155: 57. Broom 1929: 33, fig. 25H. Broom 1932: 31, fig. 6H. Boonstra 1953a: 27. Haughton & Brink 1955: 46. Boonstra 1962: 77-78, fig. 17. Comments Haughton (1915b) considered Titanosuchus to be a therocephalian. Broom (1929) placed the species in a new genus Dinocynodon. Boonstra (1953a) retained it as Titanosuchus dubius. Boonstra (1969) stated that the poor preservation of the type made the taxon identifiable only to the family Titano- suchidae. 98 ANNALS OF THE SOUTH AFRICAN MUSEUM Family Tapinocephalidae Subfamily Tapinocephalinae Pelosuchus priscus Broom, 1905 Holotype SAM-918 Material Right dentary, partial palate, femur, tibia, coracoid, scapula, rib fragments and seven vertebrae. Locality Bokfontein, Prince Albert. Collected Du Plessis and Cairncross. Generic diagnosis Jaws differ from those of therocephalians and dinocephalians in that the teeth cannot be distinguished as incisors, canines and molars. Remains or sockets of 8 teeth present, and anterior 5 larger than posterior 3; the last 3 do not seem to have distinct sockets, but are lodged in a groove. Vertebral bodies biconcave but not deeply concave. In some the transverse processes are very large and pass upwards and outwards as in some vertebrae of Belodon. In others the transverse processes pass outward and downward as in pelycosaurs. Coracoid large. (After Broom 1905: 335-336. ) Specific diagnosis As for genus. References Broom 1905: 335-336. Broom 1932: 46, fig. 3E-G. Boonstra 1955b: 186, 199, 216, fig. 7. Haughton & Brink 1955: 48. Boonstra 1956: 164-165. Von Huene 1956: 277. Orlov 1964: 256. Boonstra 1969: 42. Comments Broom (1905) provisionally placed Pelosuchus as a diaptosaurian, but in 1932 recognized it as a tapinocephalian. Boonstra (1969) proposed Pelosuchus as a junior synonym of Keratocephalus. DINOCEPHALIA TYPE MATERIAL 99 Subfamily Struthiocephalinae Struthiocephalus whaitsi Haughton, 1915 Holotype SAM-2678 Material Weathered skull without lower jaw. Locality Vivier Siding, Beaufort West. Collected Whaits. Generic diagnosis ‘Skull large; snout relatively long and slender; frontal and temporal regions not so much elevated above snout as in Tapinocephalus; eyes look forward and outward; heavy overhanging supraorbital crests; temporal fossae larger than orbits, eliptical in shape with shorter axis parallel to axis of skull; teeth weak, undifferentiated and few in number’ (Haughton 1915a: 52). Specific diagnosis As for genus. References Haughton 1915a: 52-54, pl. 10. Gregory 1926: 240, fig. 29, table 5. Broom 1932: 37, fig. 12C. Boonstra 1951: 341. Boonstra 1952a: figs 1-2. Boonstra 19525: 511. Boonstra 1952d: 248. Boonstra 1953b: 32, 46-47, 49-51. Haughton & Brink 1955: 48-49. Boonstra 1965b: 251. Comments Boonstra (1953b) provided emended generic and specific diagnoses of this taxon. Haughton & Brink’s (1955) generic diagnosis differs from that given originally by Haughton (1915a). Boonstra (1969) recognized this species as the only valid species of Struthiocephalus. Struthiocephalus akraalensis Boonstra, 1952 Holotype SAM-3719 100 ANNALS OF THE SOUTH AFRICAN MUSEUM Material Good skull without lower jaw. Locality Abrahamskraal, Prince Albert. Collected Haughton. Generic diagnosis See Haughton (1915) for diagnosis of Struthiocephalus. Specific diagnosis Compared mainly to S. whaitsi, dorsal surface of parietal wider and not pinched in but frontal narrower and dorsally excavated; postfrontal and postor- bital have larger dorsal exposure and squamosal forms a greater part of post-temporal arch, which is also of considerably stronger build. Much larger than any of the other described species. Pineal foramen lies further forward; median occipital ridge narrow and sharp and dorsal portion not broadened; postorbital bar forms at its junction with dorsal skull surface a strong ridge parallel to the mid-line and medial to this the surface of frontal and parietal hollowed out; the surface of the postorbital skull bones is coarse with deep rugae and crocodile-like pits. (After Boonstra 1952d: 247-248.) References Boonstra 1952d: 247-248, figs 1-2. Boonstra 1952f: 238. Boonstra 1953b: 32, 50. Haughton & Brink 1955: 49. Boonstra 1969: 39. Comments With regard to the material, the museum catalogue states ‘skull and skeleton’, but there are no associated postcrania in the collection and nor are any referred to in any of Boonstra’s papers. Boonstra (1953b) provided an emended specific diagnosis and later (Boonstra 1969) proposed S. akraalensis as a junior synonym of S. whaitsi. Struthiocephaloides cavifrons Boonstra, 1952 Holotype SAM-—5607 Material Skull and scapula. DINOCEPHALIA TYPE MATERIAL 101 Locality Lammerkraal, Prince Albert. Collected Haughton, 1916. Generic diagnosis Mormosaurid with skull similar to Struthiocephalus, but without nasofron- tal boss and without interorbital postrostral step; moderately wide across the postorbitals; in dorsal view the postorbitals do not form lateral border of skull (after Boonstra 1952f: 237, 240). Specific diagnosis ‘Large; snout fairly high and strong; interorbital part of skull roof concave; distinct parietal boss; skull widest at level of postorbital arches; pineal foramen lies far back’ (Boonstra 1952f: 241). References Boonstra 1952f: 237-241, figs 1-3. Boonstra 19535: 48. Boonstra 19555: 236, fig. 29a—b. Haughton & Brink 1955: 49. Boonstra 1969: 39, 55. Comments Boonstra (1969) maintained that this was a valid taxon. Struthiocephaloides duplessisi Boonstra, 1952 Holotype SAM-11693 Material Good skull. Locality Dikbome, Laingsburg. Collected Boonstra, 1946. Generic diagnosis See Boonstra (1952f) for diagnosis of Struthiocephaloides. 102 ANNALS OF THE SOUTH AFRICAN MUSEUM Specific diagnosis ‘{Skull] small; snout low and weak; interorbital skull roof convex; parietal boss not distinct; maximum width in region of temporal fossae; pineal foramen situated far anteriorly’ (Boonstra 1952f: 241). References Boonstra 19526: 509-511, figs 1-2. Boonstra 1952f: 238, 241. Boonstra 1954b: 32, 45. Haughton & Brink 1955: 49. Boonstra 1969: 39, 55. Comments Boonstra (1952b) made this specimen the type of Struthiocephalus dupless- isi sp. nov., and later (1952f) transferred the species to his new genus Stru- thiocephaloides. Boonstra (1969) considered Struthiocephaloides duplessisi to be a valid taxon. Struthionops intermedius Boonstra, 1952 Holotype SAM-11947 Material Distorted and somewhat weathered skull without lower jaw. Locality See comments. Collected Unknown. Generic diagnosis Skull fairly lightly built with slight pachyostosis; postorbital arch mod- erately strong, with no boss-like swelling of its upper part; posterodorsal circumorbital bones only slightly thickened with little overhanging of orbits; frontals excluded from orbital border; nasals and frontals only slightly thick- ened with very slight indication of a nasofrontal swelling; in sagittal plane the middle part of nasals forms concave dorsal surface and at nasofrontal junction surface moderately convex; pineal foramen penetrates small but prominent boss directed somewhat posteriorly; snout fairly long, wide and fairly high; temporal fossa medium-sized with dorsoventral diameter much greater than anteroposterior diameter; orbits large; interorbital width large; intertemporal width fairly large; intertemporal region laterally only slightly pinched in and without any indication of parietal crest; occiput inclined much anteriorly dorsoventrally; quadrates apparently only moderately shifted forward; probably DINOCEPHALIA TYPE MATERIAL 103 14-15 teeth in upper jaw; large frontals and prefrontals; parietals small; large pineal foramen situated near occipital border; postorbital apparently just meets squamosal, but parietal forms upper edge of temporal fossa (after Boonstra 1952c: 988-989). Specific diagnosis As for genus. References Boonstra 1952c: 988-989, figs 1-2. Boonstra 1952h: 246. Boonstra 1953b: 48-50, 52-53. Haughton & Brink 1955: 50. Von Huene 1956: 277. Piveteau 1961b: 277. Boonstra 1963a: 178. Boonstra 19636: 202, 204, figs 3J, 4K. Orlov 1964: 258. Boonstra 1969: 41, fig. 10K. Comments While in the museum catalogue the locality is stated to be unknown, Boonstra (1952c) stated that it was probably recovered in the vicinity of Abrahamskraal, Prince Albert. Boonstra (1952c) originaliy placed this taxon in the Mormosauridae, a view which Orlov (1964) also supported. Boonstra (19535) emended the diagnosis of this taxon. He (1969) considered this to be a valid taxon. Struthiocephalellus parvus Boonstra, 1955 Holotype SAM-5006 Material Skull, scapula, humerus, ulna, pelvis, femur, proximal portion of tibia, vertebrae, and several small, unidentified fragments. Locality See comments. Collected Haughton, 1916-17. Generic diagnosis Skull about half size of Struthiocephalus whaitsi. Weak pachyostosis with postorbital bar relatively slender and post temporal opening roomy; occiput 104 ANNALS OF THE SOUTH AFRICAN MUSEUM fairly upright and quadrate not shifted very far anteriorly; snout moderately high; no frontonasal boss. Posterior tooth crowns spatulate in outline and labiolingually compressed. In one or two of the rear crowns there appears to have been a stronger central cusp with a weaker posterior and anterior cusp as in Agnosaurus and Rhopalodon. Cervical vertebrae similar to Moschops and Moschognathus and to those of the synapsids generally. Proatlas stout. Atlas temnospondylus with paired neural arch lying on large atlantal intercentrum and odontoid-like pleurocen- trum. Atlantal intercentrum large with facet for capitulum of atlantal rib. Axial neural arch halves fused to each other and to pleurocentrum to form holo- spondylous vertebra. Spine comb-shaped. Axial centrum laterally excavated below diapophysis with sharp ventral keel. Third intercentrum smaller than its predecessors. Third cervical vertebra deeply excavated below diapophysis so that ventrally it shows a sharp keel. Atlantal rib greatly flattened, small bone with weak shaft, greatly expanded leaf-like tuberculum and much weaker capitulum. (After Boonstra 1955a: 180-184.) Specific diagnosis As for genus. References Boonstra 1955a: 180-184, figs 1-3. Boonstra 19556: 203, 225, 237-238, figs 27c, 29c, 30-33. Von Huene 1956: 277. Piveteau 1961b: 277. Orlov 1964: 258. Comments The locality, according to the museum catalogue, is Wilgebosch-Drift, Beaufort West. Boonstra (1955a, 1955b) gave it as Abrahamskraal, Prince Albert. Orlov (1964) placed this taxon into the Mormosauridae. Boonstra (1969) proposed to synonymize it with Struthiocephalus and recognized this specimen as a juvenile of S. whaitsi. Boonstra (1969) considered the material assigned to S. parvus to represent juvenile specimens of Struthiocephalus whaitsi. The legend to fig. 27c in Boonstra (1955b) erroneously reads Struthiocephalus instead of Struthiocephalellus. Subfamily Riebeeckosaurinae Riebeeckosaurus longirostris Boonstra, 1952 Holotype SAM-3400 DINOCEPHALIA TYPE MATERIAL 105 Material Incomplete skull with part of lower jaw, vertebrae and several associated, unidentified bone fragments. Locality Near Vivier Siding, Beaufort West. Collected Whaits, 1915. Generic diagnosis “Tapinocephalid with large skull, long and quite narrow; snout long, narrow and quite high; cranial bones quite thickened; postorbital arch very massive; temporal fossa roomy with dorsoventral diameter much greater than anteroposterior diameter; intertemporal area very narrow and forms sagittal crest which curves downwards and backwards; parietal forms part of supratem- poral edge; occiput very inclined; parietal very small; frontal large, does not form part of supraorbital border, but reaches supratemporal border; prefrontal forms supraorbital border; quadrate supported by anteriorly directed process of quadratojugal; teeth undifferentiated’ (Boonstra 1952h: 248). Specific diagnosis As for genus. References Boonstra 1952h: 246-249, figs 1-3. Haughton & Brink 1955: 53. Von Huene 1956: 277. Piveteau 1961b: 275. Boonstra 1963a: 178. Boonstra 19636: 200, 205, figs 3E, 4F, 6F. Orlov 1964: 258. Boonstra 1969: 42, fig. 13. Boonstra 1971: 21, 33. Comments Boonstra (1963) placed this taxon in the subfamily Riebeeckosaurinae and in 1969 he placed that subfamily within the family Tapinocephalidae. Haughton & Brink (1955) proposed that the taxon be placed in the family Moschopidae, while Orlov (1964) considered it to represent the family Mormosauridae. Boonstra (1969) considered it to be a valid taxon. 106 ANNALS OF THE SOUTH AFRICAN MUSEUM Subfamily Moschopinae Delphinognathus conocephalus Seeley, 1892 Holotype SAM-713 Material Incomplete, weathered skull with part of lower jaw. Locality Doubtful—perhaps near Beaufort West, according to Seeley (1892). Collected ?Bain, 1883. Generic diagnosis Broad, high and vertical occipital plate; large pineal foramen in middle of prominent boss (cone) with foramen at level of posterior border of orbit; orbits placed far back; quadratosquamosal region directed obliquely forward; lower jaw short and singularly deep posteriorly; occipital plate higher than wide; temporal fossa relatively small; prefrontal region concave; sub-ovate notch in inferior margin of posterior margin of posterior part of temporal arch (after Seeley 1892: 469-475). Specific diagnosis As for genus. References Seeley 1892: 469-475, figs 1-2. Broom 1910: 206, fig. 4. Broom 1914: 135-136. Gregory 1926: 228-230, 249, fig. 22a, tables I, Ill, IV. Broom 1932: 44-45, fig. 9K. Boonstra 1936: 93. Haughton & Brink 1955: 51. Von Huene 1956: 277. Boonstra 1957: 15-17, 19-21, 27, 29, 33, 36—-37, fig. 10. Boonstra 1963a: 178. Boonstra 19635: 202, 205, figs 3F, 4G. Orlov 1964: 258. Boonstra 1969: 42, 55, fig. 11c. Tatarinov 1976: 46. Comments Seeley’s (1892) diagnosis of this taxon was emended by Broom (1932), Haughton & Brink (1955) and Boonstra (1957). Gregory (1926) placed the DINOCEPHALIA TYPE MATERIAL 107 taxon in the subfamily Moschopinae, while Boonstra (1936) and Orlov (1964) were prepared only to place it in the family Moschopidae. Boonstra (1963) placed it in the family Tapinocephalidae and in 1969 proposed that the specimen be regarded as a juvenile of Moschops. Tatarinov (1976) placed the taxon in the family Delphinognathidae. Moschosaurus longiceps Haughton, 1915 Holotype SAM-3015 Material Good but weathered skull and six vertebrae. Locality La-de-da, Beaufort West. Collected Haughton, 1914. Generic diagnosis Skull long, low and narrow; parietal region slightly elevated but not tremendously thickened; nares rather far back; orbits in posterior half of skull and larger than temporal openings; quadrate in plane of middle of orbit; lower jaw massive; premaxillary teeth large with long anterior and smaller posterior cusps; postorbital bar comparatively weak; postorbital bone forms large part of upper border of temporal fossa; pineal foramen large and placed very far back (after Haughton 1915c: 78-81). Specific diagnosis As for genus. References Haughton 1915c: 78-81, figs 8-9. Broom 1923: 663. Gregory 1926: 227, 241, fig. 21, table V. Broom 1932: 45, fig. 12D. Boonstra 1936: 93-95, 97. Boonstra 1952g: 243-244. Haughton & Brink 1955: 52-53. Von Huene 1956: 276, fig. 317. Boonstra 1955a: 183. Piveteau 1961b: 277. Boonstra 1963a: 178, 188. 108 ANNALS OF THE SOUTH AFRICAN MUSEUM Boonstra 19635: 199-200, figs 2D, 3D, 4E. Orlov 1964: 258, fig. 216. Boonstra 19655: 265, fig. 11. Boonstra 1969: 39, fig. 10E. Comments Haughton (1915) considered this taxon to be titanosuchid, and Broom (1923) regarded the specimen as an immature titanosuchid. Gregory (1926) considered the taxon referrable to the family Tapinocephalidae and erected the subfamily Moschosaurinae to accommodate it. Broom (1932) agreed that it represents a tapinocephalid as did Boonstra (1936), who erected Gregory’s subfamily to familial rank, viz, Moschosauridae, and also emended the diagno- sis of the taxon Moschosaurus longiceps. In 1963, however, Boonstra (1963b) changed the rank of the family Moschosauridae back to subfamilial status. Boonstra (19655) proposed to synonymize the taxon with Struthiocephalus, a view that he maintained later (Boonstra 1969). Agnosaurus pienaari Boonstra, 1952 Holotype SAM-11832 Material Weathered, incomplete skull. Locality Lammerkraal, Prince Albert. Collected Presented by Pienaar. Generic diagnosis ‘Moschosaurus-like Tapinocephalid [sic] with slight pachyostosis; large temporal fossa; snout quite long, high and narrow; quadrate shifted forward moderately; anterior teeth typical tapinocephaloid but posterior teeth with spatulate crowns and cylindrical roots’ (Boonstra 1952g: 245). Specific diagnosis As for genus. References Boonstra 1952g: 242-245, fig. 1. Boonstra 19535: 52. Haughton & Brink 1955: 52. DINOCEPHALIA TYPE MATERIAL 109 Boonstra 1955a: 182-183. Von Huene 1956: 277. Piveteau 1961b: 277. Boonstra 1963a: 178. Orlov 1964: 258. Boonstra 1969: 42. Comments Orlov (1964) placed this taxon in the family Moschosauridae. Boonstra (1969) regarded A. pienaari as a junior synonym of Moschops. Avenantia kruisvleiensis Boonstra, 1952 Holotype SAM-9166 Material Nearly complete skull, proximal part of femur, two vertebrae, ?fibula and two unidentified fragments. See comments. Locality Kruisvlei, Beaufort West. Collected Boonstra, 1929. Generic diagnosis ‘Moschopid-like Tapinocephalid [sic], but exoccipital forms part of con- dyle, narrow intertemporal region, low parietal crest, postorbital meets squa- mosal in temporal fossa, pineal foramen surrounded by a distinct ridge’ (Boonstra 1952/7: 225). Specific diagnosis As for genus. References Boonstra 1952h: 248. Boonstra 19527: 250—255, figs 1-3. Haughton & Brink 1955: 50. Boonstra 19555: 258. Von Huene 1956: 276. Boonstra 1957: 16-22, 24-28, 33, 35-37, fig. 11. Boonstra 1963a: 178. Boonstra 1963b: 202, 205, figs. 3G, 4H. 110 ANNALS OF THE SOUTH AFRICAN MUSEUM Orlov 1964: 258. Boonstra 1969: 42, 55. Boonstra 1971: 21, 33. Comments There is some doubt regarding the postcranial material labelled SAM-9166. Specimens SAM-—9166 and SAM-—9167 are from the same locality, and Boonstra (19527) noted that SAM-—9166 is one of a ‘wagon-load of spe- cimens found partially weathered out in a small area’. The museum catalogue entry for SAM—9166 notes only ‘skull’; the entry for SAM-—9167 notes ‘ilium, femur, fibula, vertebrae’. Boonstra (1955b) mistakenly referred to SAM-—9167 as the type of Avenantia kruisvleiensis, and recorded it as having scapula, ilium, femur, and skull. The museum catalogue entry for SAM—9167, however, does not refer to a skull. It appears that the postcranial material labelled SAM—9166 may belong instead to SAM-—9167. Boonstra (1957) provided an emended diagnosis for this taxon. Orlov (1964) referred it to the family Moschopidae and Boonstra (1969) referred it to the subfamily Moschopinae as a valid taxon. Boonstra (1952h) referred to Avenantia kruisvleiensis in an article which preceded the paper (Boonstra 19527) in which the type description of the taxon was given. Accordingly, Boonstra’s (1952h) use of the name Avenantia kruis- vieiensis made it a nomen nudum. However, Avenantia kruisvleiensis was made available in his second paper (Boonstra 1952i). Moschops koupensis Boonstra, 1957 Holotype SAM-11582 Material Good skull with part of lower jaw. Locality Die Krans, Prince Albert. Collected Boonstra and Bothma, 1940. Generic diagnosis See Broom (1911) for diagnosis of Moschops. Specific diagnosis ‘(Skull] very wide across parietals with interorbital width 70 per cent of intertemporal width. Snout relatively narrow. Occipital surface greatly reduced by overgrowth from above and from the sides. Transverse pterygoidal rami are strong.’ (Boonstra 1957: 32-33.) DINOCEPHALIA TYPE MATERIAL 111. References Boonstra 1957: 18-19, 21-26, 32-33, figs 5—9. Boonstra 1969: 42. Comments Boonstra (1969) considered this specimen to represent a valid species of Moschops. Family Styracocephalidae Styracocephalus platyrhynchus Haughton, 1929 Holotype SAM-8936 Material Incomplete, crushed and badly weathered skull and part of lower jaw. Locality Boesmansrivier, Beaufort West. Collected Boonstra, 1928. Generic diagnosis Most striking features are large backwardly projecting tabular horns, massiveness of bones in postorbital region, small temporal opening, swollen cheek-like quadratojugal, and shallowness of snout which is also fairly long and narrow. Pineal foramen small and situated in middle of low, broad swelling in middle of parietal surface. Teeth present on pterygoids and palatines. (After Haughton 1929: 55-57.) Specific diagnosis As for genus. References Haughton 1929: 55-60, figs 3-5. Broom 1932: 139, fig. 47A. Boonstra 1934: 465, 467-470. Romer 1945: 601. Haughton & Brink 1955: 54. Von Huene 1956: 287-288. Heyler 1961: 127. Boonstra 1963a: 176-178, 188, figs 2L, 3J, 4H, 12. 1D ANNALS OF THE SOUTH AFRICAN MUSEUM Boonstra 1963b: 196, 199, 206, figs 1, 2, 3C, 4D, 5G, 6G. Orlov 1964: 267. Romer 1966: 372. Boonstra 1969: 44, fig. 14. Boonstra 1971: 18, 22, 24-26, 33-34, 40, 42, fig. 3. Boonstra 1972: 316, 321, 325-326, fig. 2. Tatarinov 1974: 51-52. Kitching 1977: 33. Comments Haughton (1929) referred this taxon to a new suborder, Styracocephalia. Broom (1932) placed it into another suborder, Burnetiamorpha, and consid- -ered that the specimen’s affinities were with the Gorgonopsia rather than with the Dinocephalia. Boonstra (1934) regarded the specimen as a dinocephalian and maintained it in Haughton’s Styracocephalia. Romer (1945) followed Broom, however, placing the taxon in the Gorgonopsia and in the family Burnetiidae (=Burnetiamorpha of Broom). Haughton & Brink (1955) followed Boonstra (1934) and placed it in the Dinocephalia, using Haughton’s subordinal classification of Styracocephalia. Heyler (1961) followed Romer (1945), how- ever, and placed it in the gorgonopsian family Burnetiidae. Von Huene (1956: 287) considered this taxon to belong to the Dinocephalia, but placed it into the ‘Familienkreis u. Familie Burnetiamorpha’. Boonstra (1963a) still considered it to be a dinocephalian and proposed the new family rank, Styracocephalidae for its placement. Orlov (1964) regarded it as belonging to the Burnetidae [sic] of the Gorgonopsoidea, and Romer (1966) re-affirmed his earlier (1945) convic- tion of its taxonomic placement. Boonstra (1969, 1971) proposed to maintain it in the family Styracocephalidae, which he regarded in 1972 as belonging to the Titanosuchia. Tatarinov (1974) considered that the taxon should be placed in the Burnetiidae of the order ‘Gorgonopia’, whilst Kitching (1977) maintained that Styracocephalus is a tapinocephalid and should be placed in the subfamily Tapinocephalinae. ACKNOWLEDGEMENTS We wish to thank Dr M.A. Cluver for reading and constructively criticizing this paper, and Mrs E. Blaeske and Mrs P. Eedes for typing the manuscript. REFERENCES Boonstra, L. D. 1934. On an aberrant gorgonopsian Burnetia mirabilis. S. Afr. J. Sci. 31: 462-470. BoonstrA, L. D. 1936. Some features of the cranial morphology of the tapinocephalid deinocephalians. Bull. Am. Mus. nat. Hist. 72: 75-98. BoonstTrA, L. D. 1948. Miljoene jare gelede in die Karoo. Johannesburg: Voortrekkerpers. Boonstra, L. D. 1951. Kurze Notiz tiber den Schadel der Dinocephalen-Gattung Keratoce- phalus F. v. Huene. Neues Jb. Geol. Paldont. Mh. 11: 341-343. DINOCEPHALIA TYPE MATERIAL IL) BoonstrA, L. D. 1952a. Further observations on the type skull of Struthiocephalus whaitsi. Am. Mag. nat. Hist. (12) 5: 455-459. Boonstra, L. D. 1952b. On a new tapinocephalid deinocephalian. Ann. Mag. nat. Hist. (12) 5: 509-511. Boonstra, L. D. 1952c. A new deinocephalian from the Karroo. Ann. Mag. nat. Hist. (12) 5: 988-989. Boonstra, L. D. 1952d. ’n Nuwe soort van tapinocephalide deinocephaliér: Struthiocephalus akraalensis sp. nov. S. Afr. J. Sci. 48: 247-248. Boonstra, L. D. 1952e. ’n Nuwe titanosuchiérsoort (Anteosaurus abeli). Tydskr. Wet. Kuns 12: 150-151. Boonstra, L. D. 1952f. Struthiocephaloides: ’n nuwe genus van mormosauride tapinocepha- liérs. Tydskr. Wet. Kuns 12: 237-241. BoonstrA, L. D. 1952g. Agnosaurus gen. nov.: ’n nuwe geslag van die deinocephaliérs. Tydskr. Wet. Kuns 12: 242-245. Boonstra, L. D. 1952h. ’n Nuwe tapinocephalide, Riebeeckosaurus longrostris gen. et sp. nov. Tydskr. Wet. Kuns 12: 246-249. Boonstra, L. D. 19527. ’n Uiters interessante nuwe deinocephaliér, Avenantia kruisvleiensis, gen. et sp. nov. Tydskr. Wet. Kuns 12: 250-255. BoonstraA, L. D. 1953a. A suggested clarification of the taxonomic status of the South African titanosuchians. Ann. S. Afr. Mus. 42: 19-28. Boonstra, L. D. 1953b. The cranial morphology and taxonomy of the tapinocephalid genus Struthiocephalus. Ann. S. Afr. Mus. 42: 32-53. Boonstra, L. D. 1954a. The cranial structure of the titanosuchian: Anteosaurus. Ann. S. Afr. Mus. 42: 108-148. BoonstrA, L. D. 19546. The smallest titanosuchid yet recovered from the Karroo. Ann. S. Afr. Mus. 42: 149-156. BoonstrA, L. D. 1954c. Paranteosaurus gen. nov.: a titanosuchian reptile. Ann. S. Afr. Mus. 42: 157-159. Boonstra, L. D. 1955a. Struthiocephallelus: a new deinocephalian. Ann. S. Afr. Mus. 42: 180-184. Boonstra, L. D. 1955b. The girdles and limbs of the South African Deinocephalia. Ann. S. Afr. Mus. 42: 185-326. Boonstra, L. D. 1956. The skull of Tapinocephalus and its near relatives. Ann. S. Afr. Mus. 43: 137-169. BoonstrA, L. D. 1957. The moschopid skulls in the South African Museum. Ann. S. Afr. Mus. 44: 15-38. Boonstra, L. D. 1962. The dentition of the titanosuchian dinocephalians. Ann. S. Afr. Mus. 46: 57-112. Boonstra, L. D. 1963a. Early dichotomies in the therapsids. S. Afr. J. Sci. 59: 176-195. Boonstra, L. D. 1963b. Diversity within the South African Dinocephalia. S. Afr. J. Sci. 59: 196-206. BoonstrA, L. D. 1965a. The Russian deinocephalian Deuterosaurus. Ann. S. Afr. Mus. 48: 233-236. BoonstrA, L. D. 1965b. The skull of Struthiocephalus kitchingi. Ann. S. Afr. Mus. 48: 251-265. BoonstrA, L. D. 1966. The dinocephalian manus and pes. Ann. S. Afr. Mus. 50: 13-26. Boonstra, L. D. 1969. The fauna of the Tapinocephalus Zone (Beaufort beds of the Karoo). Ann. S. Afr. Mus. 56: 1-73. BoonstrA, L. D. 1971. The early therapsids. Ann. S. Afr. Mus. 59: 17-46. Boonstra, L. D. 1972. Discard the names Theriodontia and Anomodontia: a new classification of the Therapsida. Ann. S. Afr. Mus. 59: 315-338. Broom, R. 1903. On the evidence of a new species of Titanosuchus (T. cloetei). Ann. S. Afr. Mus. 4: 142-143. Broom, R. 1904. Notice of a new fossil reptile (Scapanodon duplessisi) from the lower Karoo beds of Prince Albert. Rec. Albany Mus. 1: 182-183. Broom, R. 1905. Notice of some new fossil reptiles from the Karoo beds of South Africa. Rec. Albany Mus. 1: 331-337. Broom, R. 1909a. Notice of some new South African fossil amphibians and reptiles. Ann. S. Afr. Mus. 7: 270-278. 114 ANNALS OF THE SOUTH AFRICAN MUSEUM Broom, R. 1909b. An attempt to determine the horizons of the fossil vertebrates of the Karoo. Ann. S. Afr. Mus. 7: 285-289. Broom, R. 1910. A comparison of the Permian reptiles of North America with those of South Africa. Bull. Am. Mus. nat. Hist. 28: 197-234. Broom, R. 1911. On some new South African Permian reptiles. Proc. zool. Soc. Lond. 1911: 1073-1082. Broom, R. 1914. A further comparison of the South African dinocephalians with the American pelycosaurs. Bull. Am. Mus. nat. Hist. 33: 135-141. Broom, R. 1923. On the structure of the skull in the carnivorous deinocephalian reptiles. Proc. zool. Soc. Lond. 1923: 661-684. Broom, R. 1928. On Tapinocephalus and two other deinocephalians. Ann. S. Afr. Mus. 22: 427-438. Broom, R. 1929. On the carnivorous mammal-like reptiles of the family Titanosuchidae. Ann. Transv. Mus. 13: 9-36. Broom, R. 1932. The mammal-like reptiles of South Africa and the origin of mammals. London: H.F. & G. Witherby. -Grecory, W.K. 1926. The skeleton of Moschops capensis Broom, a dinocephalian reptile from the Permian of South Africa. Bull. Am. Mus. nat. Hist. 56: 179-252. Haucuton, S.H. 1915a. On a new dinocephalian from the Gouph. Ann. S. Afr. Mus. 12: 52-54. Haucuton, S.H. 1915b. On two new therocephalians from the Gouph. Ann. S. Afr. Mus. 12: 55-57. Haucuton, S.H. 1915c. On a new type of deinocephalian (Moschosaurus longiceps). Ann. S. Afr. Mus. 12: 78-81. HauGurton, S.H. 1929. On some new therapsid genera. Ann. S. Afr. Mus. 28: 55-78. Haucuton, S.H. & Brink, A.S. 1955. A bibliographical list of Reptilia from the Karroo beds of Africa. Palaeont. afr. 2: 1-187. HEYLER, D. 1961. Gorgonopsia. In: PIVETEAU, J. ed. Traité de Paléontologie 6: 88-139. Paris: Masson. HUuENE, F.R. von, 1956. Paldontologie und Phylogenie der Niedren Tetrapoden. Jena: Gustav Fischer. KitcuHinc, J.W. 1977. The distribution of the Karroo vertebrate fauna. Mem. Bernard Price Inst. Palaeont. Res. Univ. Witwatersrand 1: 1131. Otson, E.C. 1962. Late Permian terrestrial vertebrates. USA and USSR. Trans Am. phil. Soc. 52: 3-224. Or.ov, Y.A. 1964. Basis of palaeontology. Moscow: State Science-Technological Publ. Owen, R. 1879. Description of fragmentary indications of a huge kind of theriodont reptile (Titanosuchus ferox, Ow.) from Beaufort West, Gouph tract, Cape of Good Hope. Q. J. geol. Soc. Lond. 35: 189-198. PIVETEAU, J. 1961a. Anteosauridae. Jn: PIVETEAU, J. ed. Traité de Paléontologie 6: 84-85. Paris: Masson. PIVETEAU, J. 1961b. Tapinocephalidae. In: PiveTEau, J. ed. Traité de Paléontologie 6: 274-278. Paris: Masson. Romer, A.S. 1945. Vertebrate paleontology. 2nd ed. Chicago: University of Chicago Press. Romer, A.S. 1966. Vertebrate paleontology. 3rd ed. Chicago: University of Chicago Press. SEELEY, H.G. 1892. On Delphinognathus conocephalus (Seeley) from the middle Karoo beds, Cape Colony. Q. J. geol. Soc. Lond. 48: 469-475. TATARINOV, L.P. 1974. Theriodonts of the USSR. Moscow: NAUKA. TATARINOV, L.P. 1976. Morphological evolution of the theriodonts and the general problems of phylogenetics. Moscow: NAUKA. VAN Hoepen, E.C.N. 1916. A new Karroo reptile. Ann. Transv. Mus. 5 (supl. 3): 217. Watson, D.M.S. 1921. The bases of classification of the Theriodontia. Proc. zool. Soc. Lond. 1921: 35-98. 6. SYSTEMATIC papers must conform to the International code of zoological nomenclature (particularly Articles 22 and 51). Names of new taxa, combinations, synonyms, etc., when used for the first time, must be followed by the appropriate Latin (not English) abbreviation, e.g. gen. nov., sp. nov., comb. nov., syn. nov., etc. An author’s name when cited must follow the name of the taxon without intervening punctuation and not be abbreviated; if the year is added, a comma must separate author’s name and year. The author’s name (and date, if cited) must be placed in parentheses if a species or subspecies is transferred from its original genus. The name of a subsequent user of a scientific name must be separated from the scientific name by a colon. Synonymy arrangement should be according to chronology of names, i.e. all published scientific names by which the species previously has been designated are listed in chronological order, with all references to that name following in chronological order, e.g.: Family Nuculanidae Nuculana (Lembulus) bicuspidata (Gould, 1845) Figs 14-15A Nucula (Leda) bicuspidata Gould, 1845: 37. Leda plicifera A. Adams, 1856: 50. Laeda bicuspidata Hanley, 1859: 118, pl. 228 (fig. 73). Sowerby, 1871: pl. 2 (fig. 8a—b). Nucula largillierti Philippi, 1861: 87. Leda bicuspidata: Nicklés, 1950: 163, fig. 301; 1955: 110. Barnard, 1964: 234, figs 8-9. Note punctuation in the above example: - comma separates author’s name and year semicolon separates more than one reference by the same author full stop separates references by different authors figures of plates are enclosed in parentheses to distinguish them from text-figures dash, not comma, separates consecutive numbers Synonymy arrangement according to chronology of bibliographic references, whereby the year is placed in front of each entry, and the synonym repeated in full for each entry, is “not acceptable. In describing new species, one specimen must be designated as the holotype; other speci- mens mentioned in the-original description are to be designated paratypes; additional material not regarded as paratypes should be listed separately. The complete data (registration number, depository, description of specimen, locality, collector, date) of the holotype and paratypes must be recorded, e.g.: Holotype SAM-—A13535 in the South African Museum, Cape Town. Adult female from mid-tide region, King’s Beach Port Elizabeth (33°51’S 25°39’E), collected by A. Smith, 15 January 1973. Note standard form of writing South African Museum registration numbers and date. 7. SPECIAL HOUSE RULES Capital initial letters (a) The Figures, Maps and Tables of the paper when referred to in the text >] 6 e.g. ‘... the Figure depicting C. namacolus...’; *. ..in C. namacolus (Fig. 10)...’ (b) The prefixes of prefixed surnames in all languages, when used in the text, it not preceded by initials or full names e.g. Du Toit but A.L.du Toit; Von Huene but F. von Huene (c) Scientific names, but not their vernacular derivatives e.g. Therocephalia, but therocephalian Punctuation should be loose, omitting all not strictly necessary Reference to the author should be expressed in the third person Roman numerals should be converted to arabic, except when forming part of the title of a book or article, such as ‘Revision of the Crustacea. Part VIII. The Amphipoda.’ Te Specific name must not stand alone, but be preceded by the generic name or its abbreviation to initial capital letter, provided the same generic name is used consecutively. Name of new genus or species is not to be included in the title: it should be included in the abstract, counter to Recommendation 23 of the Code, to meet the requirements of Biological Abstracts. J. A. VAN DEN HEEVER & F. E. GRINE DINOCEPHALIA TYPE MATERIAL IN THE SOUTH AFRICAN MUSEUM (REPTILIA, THERAPSIDA) OCTOBER 1981 ISSN 0303-2515 H At AY ~f fj j ; d / ~ ’ . ‘i ta, keh he ‘OF THE SOUTH AFRICAN EUM CAPE TOWN INSTRUCTIONS TO AUTHORS 1. MATERIAL should be original and not published elsewhere, in whole or in part. 2. LAYOUT should be as follows: (a) Centred masthead to consist of Title: informative but concise, without abbreviations and not including the names of new genera or species Author’s(s’) name(s) Address(es) of author(s) (institution where work was carried out) Number of illustrations (figures, enumerated maps and tables, in this order) (b) Abstract of not more than 200 words, intelligible to the reader without reference to the text (c) Table of contents giving hierarchy of headings and subheadings (d) Introduction (e) Subject-matter of the paper, divided into sections to correspond with those given in table of contents (f) Summary, if paper is lengthy (g) Acknowledgements (h) References (i) Abbreviations, where these are numerous 3. MANUSCRIPT, to be submitted in triplicate, should be typewritten and neat, double spaced with 2,5 cm margins all round. First lines of paragraphs should be indented. Tables and a list of legends for illustrations should be typed separately, their positions indicated in the text. All pages should be numbered consecutively. Major headings of the paper are centred capitals; first subheadings are shouldered small capitals; second subheadings are shouldered italics; third subheadings are indented, shouldered italics. Further subdivisions should be avoided, as also enumeration (never roman numerals) of headings and abbreviations. Footnotes should be avoided unless they are short and essential. Only generic and specific names should be underlined to indicate italics; all other marking up should be left to editor and publisher. 4. ILLUSTRATIONS should be reducible to a size not exceeding 12 « 18 cm (19 cm including legend); the reduction or enlargement required should be indicated; originals larger than 35 x 47 cm should not be submitted; photographs should be rectangular in shape and final size. A metric scale should appear with all illustrations, otherwise magnification or reduction should be given in the legend; if the latter, then the final reduction or enlargement should be taken into consideration. All illustrations, whether line drawings or photographs, should be termed figures (plates are not printed; half-tones will appear in their proper place in the text) and numbered in a single series. Items of composite figures should be designated by capital letters; lettering of figures is not set in type and should be in lower-case letters. The number of the figure should be lightly marked in pencil on the back of each illustration. 5. REFERENCES cited in text and synonymies should all be included in the list at the end of the paper, using the Harvard System (ibid., idem, loc. cit., op. cit. are not acceptable): (a) Author’s name and year of publication given in text, e.g.: ‘Smith (1969) describes... .’ ‘Smith (1969: 36, fig. 16) describes . “As described (Smith 1969a, 1969b; eon ee ‘As described (Haughton & Broom 1927 Dy ‘As described (Haughton et al. 1927) . Note: no comma separating name aiid! year Dagination indicated by colon, not p. names of joint authors connected by ampersand - et al. in text for more than two joint authors, but names of all authors given in list of references. (b) Full references at the end of the paper, arranged alphabetically by names, chronologically within each name, with suffixes a, b, etc. to the year for more than one paper by the same author in that year, e.g. Smith (1969a, 19695) and not Smith (1969, 1969a). For books give title in italics, edition, volume number, place of publication, publisher. For journal article give title of article, title of journal in italics (abbreviated according to the World list o, scientific periodicals. 4th ed. London: Butterworths, 1963), series in parentheses, volume number, part number (only if independently paged) in parentheses, pagination (first and last pages of article). Examples (note capitalization and punctuation) BULLOUGH, W. S. 1960. Practical invertebrate anatomy. 2nd ed. London: Macmillan. FISCHER, P. —H. 1948. Données sur la résistance et de le vitalité des mollusques. J. Conch., Paris 88: 100-140. FiscHER, P.-H., DuvAL, M. & Rarry, A. 1933. Etudes sur les échanges respiratoires des littorines. Archs Zool. exp. gen. 74: 627- 634. Konn, A. J. 1960a. Ecological notes on Conus (Mollusca: Gastropoda) in the Trincomalee region of Ceylon. Ann. Mag. nat. Hist. (13) 2: 309-320. Konn, A. J. 1960b. Spawning behaviour, egg masses and larval development in Conus from the Indian Ocean. Bull. Bingham oceanogr. Coll. 17 (4): 1-51. THIELE, J. 1910. Mollusca: B. Polyplacophora, Gastropoda marina, Bivalvia. In: SCHULTZE, L. Zoologische und anthropologische Ergebnisse einer Forschungsreise im westlichen und zentralen Siid-Afrika 4: 269-270. Jena: Fischer. Denkschr. med.-naturw. Ges. Jena 16: 269-270. (continued inside back cover) ANNALS OF THE SOUTH AFRICAN MUSEUM ANNALE VAN DIE SUID-AFRIKAANSE MUSEUM Volume 86 Band October 1981 Oktober Part 4 Deel 4 o \O” Ali, sy UID Novi © CRETACEOUS FAUNAS FROM ZULULAND AND NATAL, SOUTH AFRICA ADDITIONAL OBSERVATIONS ON THE AMMONITE SUBFAMILY TEXANITINAE COLLIGNON, 1948 By WILLIAM JAMES KENNEDY HERBERT CHRISTIAN KLINGER & HERBERT SUMMESBERGER Cape Town Kaapstad The ANNALS OF THE SOUTH AFRICAN MUSEUM are issued in parts at irregular intervals as material becomes available Obtainable from the South African Museum, P.O. Box 61, Cape Town 8000 Die ANNALE VAN DIE SUID-AFRIKAANSE MUSEUM word uitgegee in dele op ongereelde tye na gelang van die beskikbaarheid van stof Verkrygbaar van die Suid-Afrikaanse Museum, Posbus 61, Kaapstad 8000 OUT OF PRINT/UIT DRUK iL, Aas 5-8), HUD, 4-5, Biot), SEB), S, 79), GC, tapi) 7a sh 92) 7)e 101-3), VLD Sag) esa), CES), WO), i, ES). 26), 28, 250) EDITOR/REDAKTRISE Ione Rudner Copyright enquiries to the South African Museum Kopieregnavrae aan die Suid-Afrikaanse Museum ISBN 0 86813 025 7 Printed in South Africa by In Suid-Afrika gedruk deur the Rustica Bressa, Piya, etd: Die Rustica-pers, Edms., Bpk., Court Road, Wynberg, Cape Courtweg, Wynberg, Kaap CRETACEOUS FAUNAS FROM ZULULAND AND NATAL, SOUTH AFRICA ADDITIONAL OBSERVATIONS ON THE AMMONITE SUBFAMILY TEXANITINAE COLLIGNON, 1948 By WILLIAM JAMES KENNEDY Geological Collections, University Museum, Oxford HERBERT CHRISTIAN KLINGER South African Museum, Cape Town & HERBERT SUMMESBERGER Naturhistorisches Museum, Vienna (With 27 figures) [MS accepted 11 June 1981] ABSTRACT Data supplementary to the monograph on the ammonite subfamily Texanitinae Collignon, 1948, by Klinger & Kennedy are given. These include a redescription and illustration of the type material and other Austrian and French specimens of Ammonites serratomarginatus Redtenbacher, 1873, and Ammonites quinquenodosus Redtenbacher, 1873. New material permits a study of the intraspecific variation in Reginaites zulu Klinger & Kennedy, 1980, and the homoeomorphy between it and Plesiotexaniies (P.) stangeri (Baily, 1855). Two micromorph specimens of Submortoniceras woodsi (Spath, 1921) are homoeomorphic with Protexanites (P.) bontanti shimizui Matsumoto, 1970, exhibit traces of colour banding, and raise the possibility of sexual dimorphism in the subfamily Texanitinae. CONTENTS PAGE MMtTOGU CHO MEA ee Oa Meo eee ae el. sah, geben lo 116 I OCAUOMONSMECITIENS Gan ane ae aa ae seo 116 DIMENSIONSLOL SPECIMENS ase eee ee 116 Suture tenmimologyen. ase. eae ee eae os 116 ANT NOUAINON OL CHEMIE, goaecnscascuncoennsonc 116 Systematic palacontologyanna ae sae ane ese IL7/ On Ammonites serratomarginatus Redtenbacher, 1873 and Ammonites quinquenodosus Red- tenbachen el S/Se irre eee ne re ae 117 Intraspecific variation in Reginaites zulu Klinger & Kennedy, 1980, and homoeomorphy with Plesiotexanites (P.) stangeri (Baily, 1855). ... 137 Colour banding in micromorphs of Submortoni- ceras woodsi (Spath, 1921) and their homoeo- morphy with Protexanites (P.) bontanti shimi- Zuieiatsumotow 9/70 tee ee eee eee ae 149 Acknowledgements: 25. 25.4 hos wees gues oe 153, RE LERET COS NG rico Sines ead a econ tae a rae 153 ALS Ann. S. Afr. Mus. 86 (4) 1981: 115-155, 27 figs. 116 ANNALS OF THE SOUTH AFRICAN MUSEUM INTRODUCTION Subsequent to Klinger & Kennedy’s (1980a) monographical description of the South African representatives of the ammonite subfamily Texanitinae Collignon, 1948, additional data from South Africa and Europe became avail- able which are supplementary to the overall account already published. Part of these were published separately (Klinger & Kennedy 1980b). Although these data deal with different and remotely related aspects of the subfamily Texaniti- nae, economic measures deemed it advisable to publish these under one cover with appropriate subtitles. Authorship according to seniority is given below each subtitle. Authorship for the whole publication is in alphabetical order. LOCATION OF SPECIMENS The following abbreviations are used to indicate the repositories of the material studied: FSR Faculté des Sciences, Rennes, France GBA Geologische Bundesanstalt, Vienna, Austria LL Oberosterreichisches Landesmuseum, Linz, Austria NHMW_ Naturhistorisches Museum, Vienna, Austria NS Haus der Natur, Salzburg, Austria SAM South African Museum, Cape Town, South Africa SAS Geological Survey of South Africa, Pretoria DIMENSIONS OF SPECIMENS Dimensions of specimens are given in millimetres; abbreviations are as follows: D = diameter, Wb = whorl breadth, Wh = whorl height, U = umbilical dia- meter, R = ribs per whorl, U : E tbs = ratio of umbilical to external tubercles. Figures in parentheses are dimensions as a percentage of total diameter. SUTURE TERMINOLOGY The suture terminology of Wedekind (1916; see Kullmann & Wiedmann 1970 for a recent review) is followed in the present work: I = internal lobe, U = umbilical lobe, L = lateral lobe, E = external lobe. ANNOTATION OF ORNAMENT This follows the scheme presented in Klinger & Kennedy (1980a: 3-4); tubercles are numbered in ontogenetic sequence from the umbilical wall to the venter; e.g. umbilical tubercle is (1), lateral tubercle (2), submarginal (3), marginal (4), and external (5). CRETACEOUS FAUNAS FROM SOUTH AFRICA £17 SYSTEMATIC PALAEONTOLOGY ON Ammonites serratomarginatus REDTENBACHER, 1873, AND Ammonites quinquenodosus REDTENBACHER, 1873. (By W. J. Kennedy, H. Summesberger & H. C. Klinger) Revision of the texanitid faunas of the South African Upper Cretaceous (Klinger & Kennedy 1980a) raised a number of questions, amongst others the effect of differential preservation and geographic morphological variation in Ammonites serratomarginatus and Ammonites quinquenodosus, two species from the Gosau Beds of the Austrian Alps described almost a century ago by Anton Redtenbacher (1873). It has now proved possible to locate some of the type material of these two species, as well as a series of other Austrian specimens. Paratexanites serratomarginatus (Redtenbacher, 1873) Figs t=7 Ammonites serrato-marginatus Redtenbacher, 1873: 110, pl. 25 (fig. 2a—d) Mortoniceras serrato-marginatum (Redtenbacher): De Grossouvre, 1894: 69, pl. 16 (fig. la—b). Pervinquiére, 1907: 242. Diener, 1925148. Bevahites (Parabevahites) serrato-marginatus (Redtenbacher): Collignon, 1948: 83(38). Non Parabevahites serrato-marginatus (Redtenbacher): Collignon, 1966: 76, pl. 486 (figs 1962-1963). Non Parabevahites cf. serrato-marginatus (Redtenbacher): Collignon, 1966: 80, pl. 488 (figs 1967-1969). Paratexanites (Parabevahites) serratomarginatus (Redtenbacher): Matsumoto, 1970: 260, pl. 36 (figs 1-3), text-fig. 16. Matsumoto & Hirano, 1976: 337, text-fig. 3. Paratexanites (Parabevahites) serratomarginatus grossouvrei Matsumoto, 1970: 263. Paratexanites (Parabevahites) cf. serratomarginatus (Redtenbacher): Kennedy & Kollmann, 1977: 414, pl. 1 (fig. la—c). Types Redtenbacher based this species on seven syntypes, all of which have been traced. NS 6381, the original of Redtenbacher 1873, pl. 25 (fig. 2a—b) is the lectotype; paralectotypes are NS 6376, 6379, 6387, 6384, 6385, 6387. Four further specimens may also be part of the type series: NS 18427-18428 (part and counterpart), NS 18435, LL 26590, and 26598, all from the Coniacian Gosau Beds of Glanegg, Austria. Other specimens studied NHMW 1978/2029/3 from the Coniacian Beds of Glanriedel, Austria, FSR ‘A’, Seunes Collection, FSR 2834 and 2835, all from the Coniacian Craie de Villedieu of La Ribochére, commune de Couture, Loir-et-Cher, France. Dimensions D Wb Wh Wb:Wh U R NS 6381 45,2 16,2(38) 14,0(33) 1,15 15,0(35) 38-39? FSR 2834 58,5 = _ = 23,0(39) 21 ESR A’ 723 — 24,5(34) — 29,8(41) 27 118 ANNALS OF THE SOUTH AFRICAN MUSEUM Description The lectotype (Figs 1A—C, 21, 3B) is a small, distorted, composite internal mould just over 42 mm in diameter. Coiling is evolute, only 25 per cent of the previous whorl being covered. The umbilicus is of moderate width (c. 35 per cent of diameter?) with a low, rounded wall. The whorl section is slightly depressed (due to post-mortem crushing), with greatest breadth below mid-flank. Ornament consists of numerous slightly flexed, crowded prorsiradiate ribs, approximately 40 per whorl. Most arise singly from bullae (1) at the umbilical shoulder (although a few arise in pairs or are intercalated), and bear small, weakly clavate submarginal (3) and closely spaced, stronger marginal (4) tubercles. There are stronger external (5) clavi on either side of a narrow keel flanked by shallow sulci. The earliest ontogenetic stages are seen in NS 6384 (Fig. 2G—H) and LL 26598 (Fig. 2A-B). Here, ribbing is again crowded and flexuous, with the distinctive close spacing of tubercles 3 + 4 visible even at a whorl height of only 5,5 mm. The former specimen has a compressed whorl section, presum- ably due to crushing, but reveals a better preserved keel than the lectotype, and this shows unmistakable undulations corresponding to the external (5) clavi. This is also seen, but is less prominent, in LL 26598, which differs from the lectotype in having the submarginal (3) tubercle longer and more pro- minent than the marginal (4) (Fig. 2A—B). Of the specimens corresponding in size to the lectotype, NS 18428 (NS 18427 is the counterpart) (Figs 1D, 3C) is coarser ribbed, with stronger bullae (1) and a greater forward projection to the ventrolateral ribbing. This is also seen in NS 6379 (Fig. 11). In contrast, NS 6382 (Figs 1K, 3D), although crushed, is as densely ribbed as the lecto- type. Of the larger specimens, NS 6387, is the original of Redtenbacher 1873, pl. 25 (fig. 2C—D). It has a maximum whorl height of 20,5 mm, and appears to be sparser ribbed than the lectotype, as do all the larger specimens, suggesting outer whorls were consistently more distantly ribbed than the nuclei (Figs 1F-G, 3A). The ribs are flexuous with the submarginal (3) and marginal (4) tubercles close together, both clavate, and the former weaker than the latter (Figs 1G, 3A). The keel is well preserved, and feebly undulose. NS 6376 is identical, as far as it is preserved (Fig. 1E). In contrast, NS 18453 (Fig. 1H) shows the submarginal (3) and marginal (4) tubercles to be relatively stronger in relation to the ribs, with the submarginal (3) far more prominent, although shorter than the marginal (4). The largest and most ontogenetically advanced fragment, LL 26590 (Fig. 2C—D) shows a distinctly undulose keel and submar- ginal (3) and marginal (4) tubercles grouped on a swelling on the ventrolateral shoulder, with the external (5) clavi seemingly linked by weak ridges, which are much accentuated by crushing (Fig. 2D). None of these specimens shows the sutures. CRETACEOUS FAUNAS FROM SOUTH AFRICA 119 Fig. 1. Paratexanites serratomarginatus (Redtenbacher). A—-C. The lectotype, NS 6381, the original of Redtenbacher (1873), pl. 25 (fig 2a—b). D. Paralectotype, NS 18428. E. Paralecto- type, NS 6376. F—G. Paralectotype, NS 6378, the original of Redtenbacher (1873), pl. 25 (fig. 2c—d). H. Paralectotype, NS 18453. I. Paralectotype, NS 6379. J. Paralectotype, NS 6385. K. Paralectotype, NS 6382. All specimens are from the Coniacian Gosau Beds of Glanegg, Austria. X 1. 120 ANNALS OF THE SOUTH AFRICAN MUSEUM G H | Fig. 2. Paratexanites serratomarginatus (Redtenbacher). A—-B. Paralectotype, LL 26598. C—D. Paralectotype, LL 26590. E-F. FSR 2835. G—H. Paralectotype, NS 6384. I. Lectotype, NS 6381. A-D, G-I are from the Coniacian Gosau Beds of Glanegg, Austria; E-F is from the Coniacian Craie de Villedieu of La Ribochére, Loir-et-Cher, France. A-B, I X 2; C-D, E-F x 1; G-H x 4. CRETACEOUS FAUNAS FROM SOUTH AFRICA 124 Fig. 3. Paratexanites serratomarginatus (Redtenbacher). A. Paralectotype, NS 6387. B. Lecto- type, NS 6381. C. Paralectotype, NS 18428. D. Paralectotype, NS 6382. All specimens are from the Coniacian Gosau Beds of Glanegg, Austria. x 2. 122 ANNALS OF THE SOUTH AFRICAN MUSEUM Discussion The above description, based entirely on material from Glanegg, shows this species to be variable in both density and strength of ornament. A small, compressed specimen from Glanriedel, Austria, NHMW 1978/2029/3 (Fig. 4), is comparable to material from Glanegg. De Grossouvre (1894: 69, pl. 16 (fig. la—b) refigured herein as Fig. 7A—B) described a much larger specimen from the Craie de Villedieu of La Ribochére, Couture, Loir-et-Cher, France, which Matsumoto (1970: 263) has designated holotype of the subspecies P. serrato- marginatus grossouvrei. Matsumoto (1970: 262) differentiated it from the Austrian form because of the ‘stronger and more rounded inner ventrolateral G H Fig. 4. Paratexanites serratomarginatus (Redtenbacher) NHMW 1978/2029/3 from the Coniacian Gosau Beds of Glanriedel, Austria. A-D Xx 1; E-H x 2. CRETACEOUS FAUNAS FROM SOUTH AFRICA 123 Fig. 5. Paratexanites serratomarginatus (Redtenbacher). A-C. FSR ‘A’, Seunes Collection. D-F. FSR 2834. Both specimens from the Coniacian Craie de Villedieu of La Ribochére, Loir-et-Cher, France. X 1. 124 ANNALS OF THE SOUTH AFRICAN MUSEUM tubercles which are approximated to the outer ones, forming double ventrolat- eral major protruberances . . . the ventral keel shows weak undulations which correspond in number to the radial ribs, but in the Alpine species the keel is continuous and not undulated’. As the description of the Austrian material shows, the criterion of undulose v. continuous keel is not valid. Equally our largest and ontogenetically most advanced specimen (Fig. 2C-D) shows the double protuberance of the holotype of P. serratomarginatus grossouvrei, and there is such variation in the relative development of submarginal (3) and marginal (4) tubercles in the Austrian type series that this scarcely seems a criterion for separation. The authors figure, however, three smaller French specimens from the Coniacian part of the Craie de Villedieu of La Ribochére (Figs 2E-F, SA-F, 6) that show a greater range of variation and more advanced ontogenetic develop- Us Fig. 6. Paratexanites serratomarginatus (Redtenbacher). External suture of FSR ‘A’, Seunes Collection. x 6. ment than is seen in the Austrian specimens. All are sparser ribbed than the lectotype, some with only half as many ribs at a somewhat greater diameter, although there is overlap in rib density with other types. FSR ‘A’ (Fig. SA—C) is very close indeed to the Austrian type material, but FSR 2835 (Fig. 2E-F) is a curious, slowly expanding form with fewer ribs, which are more flexuous and have stronger bullae, with the marginal (4) tubercle very small indeed (Fig. 2E). FSR 2834 (Fig. 5D-F) shows similar features, especially the prominence of the submarginal (3) tubercle, while the marginal (4) tubercle is almost indistin- guishable at the beginning of the last whorl, although strengthening towards the aperture. Given this variation and the overlap between the Austrian and 5 CRETACEOUS FAUNAS FROM SOUTH AFRICA ‘60 x poonpoy *(q-e] “3y) OT ‘Id ‘p6gT o1ANOssoIyH oq Jo Adod ‘soue14 ‘19YD-19-IOT ‘o1gYydoqny eT JO noIpal|iA op JVI URIORIUOD IY} WIJ ‘OJOUUNSJRI] 1a4ANOssos (1OYeQUa\pIy) SNJOUIsADWOJDAAaS SajluDxXavADg JO ddAjojoy 94, *L “314 126 ANNALS OF THE SOUTH AFRICAN MUSEUM French specimens, we doubt that they merit subspecific separation, although acknowledging differences in variation range. Specimens described from Japan by Matsumoto (1970: 260, pl. 36 (figs 1-3), text-fig. 16) fall into P. serratomarginatus as here interpreted. Collignon’s (1966) specimens have a lateral tubercle and should be referred to Plesiotexan- ites, while we would continue to refer to the Zululand material (Klinger & Kennedy 1980a: 59, figs 45-47A) as Paratexanites sp. aff. P. serratomarginatus. When compared to other ‘Parabevahites’, Paratexanites emscheris (Schliiter, 1876: 155, pl. 42 (figs 8-10)) has coarser, straighter ribs and umbilical bullae low on the flank rather than perched on the umbilical shoulder, and there are other differences as noticed by Matsumoto (1970: 262). Paratexanites sellardsi Young (1963: 79, pl. 32 (fig. 7), pl. 36 (figs. 3-5), pl. 37 (fig. 1), pl. 39 (fig. 4), pl. 49 (fig. 3), text-fig. 17) is also more coarsely ribbed and robustly tuberculate. Occurrence The type specimens are from Glanegg, Austria. Other ammonites from this locality (Brinkmann 1935: 2; Kennedy & Summesberger, in preparation) include Gaudryceras glaneggense (Redtenbacher), Otoscaphites arnaudi (De Grossouvre), ‘Ammonites’ aberlei Redtenbacher, Gauthiericeras margae (Schliiter), and Protexanites sp. juv. These suggest a stratigraphic level way above the base of the Coniacian stage. The species is also recorded from France, Tunisia and Zululand and, where reliably dated, is also of Coniacian age. Texanites quinquenodosus (Redtenbacher, 1873) Figs 8-16 Ammonites texanus von Hauer (non Roemer, 1852) 1858: 10, pl. 2 (figs 4-6). Ammonites quinquenodosus Redtenbacher, 1873: 108, pl. 24 (fig. 3a—b). Mortoniceras texanus Schliter sp.: Zurcher, 1905: 686. Mortoniceras quinquenodosum (Redtenbacher): Yabe & Shimizu, 1923: 30 (partim). Mortoniceras quinquenodosusm [lapsus] (Redtenbacher): Diener, 1925: 147 (partim). ?Texanites quinquenodosus var. evoluta Haas, 1942: 18, text-fig. 12. Texanites quinquenodosus (Redtenbacher): Collignon, 1948: 69, text-fig. 2; 71966: 128, pl. 510 (fig. 2021). Thomel, 1969: 12, pl. G (figs 1-2). Klinger & Kennedy 1980a: 135, figs 102-3. Texanites (Texanites) quinquenodosus (Redtenbacher): Matsumoto, 1970: 272. Wiedmann, 1979: 48, pl. 7 (figs C-D). Texanites aff. quinquenodosus (Redtenbacher): Fabre-Taxy, 1963: 17, pl. 1 (fig. 14). Texanites (Texanites) sp. aff. T. (T.) quinquenodusus (Redtenbacher): Matsumoto, 1970: 273, pl. 42 (fig. 3a—b), pl. 46 (figs 1-3). Type The lectotype is GBA 1873/01/13, the original of Redtenbacher 1873, pl. 24 (fig. 3a—b), from the Santonian Gosau Beds of St Wolfgang, Ischl, Austria. CRETACEOUS FAUNAS FROM SOUTH AFRICA 127, Other specimens studied In addition to the lectotype, the following specimens were available: LL 41/1938, a paralectotype, from the Schneiderwirtsbrticke, Ischl; LL 32 from Gosau; GBA _ 1873/01/13 (SV 2569), a paralectotype from Gosau; NHMW 1935.11.39, from Grabenbach, Gosau; NHMW 1926.II.2469, from the Nef- graben, Russbach, Gosau; NHMW 1935.III.40, from Gosau; GBA, an unregis- tered fragment from Gosau, and a possible paralectotype; NHMW 1935.III.41, from Gosau; and Bohm Collection from the Nefgraben, Gosau. All specimens are of Santonian age. Dimensions D Wb Wh Wb:Wh U R NHMW 1935.11.40 118,5 32(27) 40(34) 0,8 51(43) 27 Description The lectotype (Fig. 8) is a flattened and distorted internal mould with almost a whorl of body chamber preserved, and a maximum diameter of almost 160 mm. Coiling is very evolute, the inner whorls being exposed to the outer ventrolateral (4) tubercle, which is housed in a small notch in the umbilical wall of the succeeding whorl. In the best preserved specimen, NHMW 1935.III.40 (Figs 12-13, 14A), the broad, shallow umbilicus comprises 43 per cent of the diameter. Uncrushed specimens show the intercostal whorl section to have been high oval with convergent flanks and a breadth to height ratio of 0,75. The costal section is also compressed, with the maximum width at the lateral (2) tubercle, and a breadth to height ratio of 0,87. Ornament consists of numerous (27-32) straight, simple, recti- to slightly prorsiradiate ribs. These arise at the umbilical seam, and bear a pointed umbilical (1) bulla directed into the umbilicus. There are a larger, feebly clavate lateral (2) tubercle, a larger, clavate submarginal (3) tubercle and a strong clavate marginal (4) tubercle, and a long external (5) tubercle, which tends to link with its neighbours into an undulose keel. There is a strong, faintly undulose siphonal keel flanked by distinct grooves (Fig. 14A-B). On adult body chambers (e.g. LL 41/1938 (Figs 15-16)), the tubercles decline and the ribs crowd towards the aperture. The suture line is quite simple (Fig. 11) with moderately incised elements. Discussion Specimens available vary somewhat in strength and number of ribs while, when shell is preserved, the tubercles are also much sharper than on moulds during early growth (Figs 9-10), and there are strong transverse growth striae in adults (Figs 15-16). Matsumoto (1970: 273, pl. 42 (fig. 3), pl. 46 (figs 1-3)) has described a series of specimens from the Lower Santonian of Hokkaido, Japan, which shows developmental stages at much smaller sizes than seen in the Austrian 128 ANNALS OF THE SOUTH AFRICAN MUSEUM SS SAe: Fig. 8. Texanites quinquenodosus (Redtenbacher). The lectotype, GBA 1873/01/13, the orig- inal of Redtenbacher 1873, pl. 24 (fig. 3a) from the Santonian Gosau Beds of St Wolfgang, ! Austria. X 1. CRETACEOUS FAUNAS FROM SOUTH AFRICA 129 Fig. 9. Texanites quinquenodosus (Redtenbacher). Paralectotype, GBA 1873/01/13 (SV 2569), the original of Redtenbacher 1873, pl. 24 (fig. 3a—b) from the Santonian Gosau Beds of St Wolfgang, Austria. A-C x 1; D x 2. 130 ANNALS OF THE SOUTH AFRICAN MUSEUM C Fig. 10. Texanites quinquenodosus (Redtenbacher). A-C. NHMW 1935. III. 41, from the Santonian Gosau Beds of Gosau, Austria. x 1. CRETACEOUS FAUNAS FROM SOUTH AFRICA 131 Fig. 11. Texanites quinquenodosus (Redtenbacher). External suture of GBA 1873/01/13 (SV2569). x 6. material. The large specimens from Japan are said to differ from the lectotype in having the ventral (=external (5)) clavus displaced in front of the outer ventrolateral, but this is seen to some degree in GBA 1873/01/13 (SV 2569) (Fig. 9A-B, D), and is probably of limited significance. Texanites quinquenodosus most closely resembles Texanites collignoni Klinger & Kennedy, 1980a ( = Texanites oliveti non Blanckenhorn sensu Col- lignon 1948, 1966; Matsumoto & Ueda 1962; Pop & Szasz 1973; Matsumoto 1978) from the Lower Santonian of Madagascar, Zululand, Japan, Romania, and possibly North America. The two species have comparable dimensions and rib densities on the outer whorls, but differ mainly in details of ornamentation. In T. quinquenodosus tuberculation is distinctly clavate, especially in the early stages of growth, whereas 7. collignoni has more spinose ornament throughout. Furthermore, crowding of ribbing and decline of tuberculation towards the body chamber is not as common in T. collignoni as in T. quinquenodosus. Given more material with precise stratigraphic data, and taking the effects of differential preservation and intraspecific variation into account, specific sepa- ration of the two may prove unnecessary. For the present, however, the clavate tuberculation in T. quinquenodosus is regarded as being sufficient to separate it from T. collignoni. IBZ ANNALS OF THE SOUTH AFRICAN MUSEUM Fig. 12. Texanites quinquenodosus (Redtenbacher). NHMW 1935. III.40, from the Santonian Gosau Beds of Gosau, Austria. xX 1. CRETACEOUS FAUNAS FROM SOUTH AFRICA 133 Fig. 13. Texanites quinquenodosus (Redtenbacher). NHMW 1935.III.40, from the Santonian Gosau Beds of Gosau, Austria. X 1. 134 ANNALS OF THE SOUTH AFRICAN MUSEUM A B Fig. 14. Texanites quinquenodosus (Redtenbacher). A. NHMW 1935.III.40, from the Santonian Gosau Beds of Gosau. B. Paralectotype, LL 41/1938, from Schneiderwirtsbrticke Ischl. A. X 1; B X 0,75. CRETACEOUS FAUNAS FROM SOUTH AFRICA 135 Fig. 15. Texanites quinquenodosus (Redtenbacher). Paralectotype LL 41/1938, from the Santonian Gosau Beds of Schneiderwirtsbriicke, Ischl. x 0,75. 136 ANNALS OF THE SOUTH AFRICAN MUSEUM Fig. 16. Texanites quinquenodosus (Redtenbacher). Paralectotype, LL 41/1938, from the Santonian Gosau Beds of Schneiderwirtsbriicke, Ischl. x 0,75. CRETACEOUS FAUNAS FROM SOUTH AFRICA ey Texanites oliveti (Blanckenhorn, 1905), according to the interpretation of Klinger & Kennedy (1980a), is a distinctly compressed, Upper(?) Santonian species. T. quinquenodosus may be separated from Texanites texanus texanus (Roemer) (see Young 1963: 80, pl. 38 (figs 1-2), pl. 40 (figs 1-3), pl. 41 (fig. 4), text-figs 21g, 22a, 25d) by its more evolute coiling and denser ribbing; these also distinguish it from T. texanus twiningi Young (1963: 82, pl. 38 (fig. 5), pl. 39 (fig. 1), pl. 41 (figs 2,5), pl. 48 (fig. 4)). Texanites texanus gallicus Collignon (1948: 75, pl. 8 (fig. 1-1a); De Grossouvre 1894: 80, pl. 16 (figs 2, 3a—b, 4a—b), pl. 17 (fig. 1)) is less compressed, and has the inner three rows of tubercles rounded. In T. texanus hispanicus Collignon (1948: 76, text-fig. 11-1la, pl. 8 (fig. 2-2b)) there are comparable differences, and the lateral tubercle migrates ventrally until it reaches a mid-flank position. The Angolan Texanites quinquenodosus evolutus Haas (1942: 18, text-fig. 12) has an umbilical diameter of 53 per cent compared to 43 in the best preserved Gosau specimen, and can only very doubtfully be referred to Redtenbacher’s species. Similarly, the specimen described as T. quinquenodo- sus from Madagascar (Collignon 1966: 128, pl. 510 (fig. 2021)) is a doubtful representative by virtue of its tighter coiling and more compressed whorl section. Occurrence The Austrian specimens can be dated no more precisely than Santonian in our present state of knowledge. The species is also recorded from the Lower Santonian of south-eastern France, Hokkaido (Japan) and, doubtfully, from the Santonian of Angola and Upper Santonian of Madagascar. INTRASPECIFIC VARIATION IN Reginaites zulu KLINGER & KENNEDY, 1980, AND HOMOEOMORPHY WITH Plesiotexanites (P.) stangeri (BAILY, 1855) (By H. C. Klinger & W. J. Kennedy) In the authors’ (Klinger & Kennedy 1980a) description of the South African representatives of the ammonite subfamily Texanitinae, two new Species definitely referable to the genus Reginaites Reyment, 1957, were erected: Reginaites zulu and Reginaites reymenti. Recent collecting at the type locality of R. zulu yielded additional material that allows the authors to comment on the intraspecific variation in the species and firmly to date it as Upper Santonian on the basis of its association with Inoceramus (Cordiceramus) muelleri Petrascheck. It also demonstrates a strik- ing homoeomorphy with the partly contemporary (though apparently geogra- phically separated) species Plesiotexanites (P.) stangeri (Baily, 1855). The 138 ANNALS OF THE SOUTH AFRICAN MUSEUM material further supports the authors’ inclusion of Reginaites in the subfamily Texanitinae, rather than in the subfamily Peroniceratinae Hyatt, as advocated by Matsumoto (1965: 238) and Wiedmann (1978: 670). Reginaites zulu Klinger & Kennedy, 1980 Figs 17-25 Reginaites zulu Klinger & Kennedy, 1980a: 116, figs 90-92, 93A-B, 94. Types The holotype by original designation is SAS-H126A/9 from Bed A at locality 105 on the south-eastern shores of False Bay, Zululand, St Lucia Formation, Upper Santonian. Paratype SAS—H126A/4 is from the same hori- zon and locality. Material Five specimens, SAM-—PCZ5952-6, all from the same locality and horizon as the types. Dimensions D Wb Wh Wb:Wh U R PCZ5955 75,0. _-25.03,3) 26,0(34,7) 0,96 — 36,0(48,0) 18 at 134,0 34,0(25,4) 35,0(26,1) 0,97 74.0(55,2) alee PCZ5953 120,0 — 32,0(26,7) 34,0(28,3) 0,94 ~—-62,0(51,7) 23 H126A/9 140,0 — 38,7(27,6) 40,0(28,6) 0,96 74,0(52,9) 21 PCZ5956 195,0 == 56,0(28,7) — oes i Description Coiling is very evolute with an umbilical diameter varying between 48 and 55 per cent, increasing with growth. The, whorl section changes through ontogeny, being subsquare on the inner whorls, becoming compressed subrect- angular with a fastigiate venter, and then distinctly compressed with a conspicu- ous central keel and subordinate lateral keels (Fig. 17A). None of the specimens has the innermost whorls preserved, but in SAM-PCZ5953 (Fig. 18) ornament at approximately 40 mm diameter already consists of pinched umbilical (1) bullae situated at the umbilical edge, and broad, low, prorsiradiate ribs bearing very large, bullate to clavate elongated ventrolateral (3 + 4) tubercles situated a little distance away from the umbilical seam of the succeeding whorl. The venter at this stage is already distinctly tricarinate, with the central keel strongest. Strength and density of ribbing is extremely variable, as the Figures show. Division of the ventrolateral tubercle takes place at varying diameters. In SAM-PCZ5953 (Fig. 18) definite signs of division are already present at a diameter of approximately 50 mm, whereas in the holotype this becomes noticeable only at a diameter of 65 mm (Figs 19-20). With increasing diameter CRETACEOUS FAUNAS FROM SOUTH AFRICA 139 A B Fig. 17. A. Reginaites zulu Klinger & Kennedy, 1980, illustrating ontogenetic changes in whorl section (after Klinger & Kennedy 1980a, fig. 90A). B. Plesiotexanites (P.) stangeri (Baily, 1855) illustrating ontogenetic changes in whorl section (after Matsumoto 1970, text-fig. 23b). All X 1. this separation becomes more obvious, resulting in a clavate marginal (4) tubercle, situated slightly forward of a larger, rounded submarginal (3) tubercle (Figs 20A-B, 21C—D). Concurrent with division of the ventrolateral tubercle, all lateral ornament weakens, the ratio of whorl breadth to whorl height decreases, and the venter becomes distinctly fastigiate. Differences in strength of ornament and stage at which attenuation takes place is variable. SAM-PCZ5956 (Fig. 22) shows attenuation of ornament and change in whorl section to the point where the flanks of the last part of the phragmocone are smooth. On body chambers (Figs. 21A-B herein) (Klinger & Kennedy 1980a, fig. 94A-B) lateral ornament is reduced to indistinct broad, low swellings. 140 ANNALS OF THE SOUTH AFRICAN MUSEUM Fig. 18. Reginaites zulu Klinger & Kennedy, 1980. SAM—PCZ5953 from locality 105, St Lucia Formation, Zululand. x 1. 141 CRETACEOUS FAUNAS FROM SOUTH AFRICA Dp 8 Fig. 19. Reginaites zulu Klinger & Kennedy, 1980. The holotype, SAS—H126/9, from locality 105, St Lucia Formation, Zululand. x 1. ANNALS OF THE SOUTH AFRICAN MUSEUM 142 The holotype, Zululand. x 1. b) 1980 ion Format 1a & Kennedy, St Luc inger ty 105, 1 tes zulu Kl inal 20. g Reg SAS-H126/9, from local Fi CRETACEOUS FAUNAS FROM SOUTH AFRICA 143 Fig. 21. Reginaites zulu Klinger & Kennedy, 1980. A-B. SAM-—PCZ5952, a body chamber fragment illustrating total disappearance of ornament. C-D. SAM-PCZ5954 illustrating division of ventrolateral tubercle. Both specimens from locality 105, St Lucia Formation, Zululand. x 1. 144 ANNALS OF THE SOUTH AFRICAN MUSEUM Fig. 22. Reginaites zulu Klinger & Kennedy, 1980. SAM—PCZ5956 illustrating ontogenetic change in ornament. From locality 105, St Lucia Formation, Zululand. x 0,5. None of the specimens shows distinct lateral (2) tubercles at any stage of ontogeny, although SAM-—PCZ5955 (Fig. 23) has faint swellings on the central part of the ribs on the inner whorls. Preservation is poor, and we cannot confirm that these are incipient lateral (2) tubercles or not. In only three specimens are part of the suture exposed: it is rather simple (Fig. 24) but varies considerably in details and the width of the lateral lobe (L). Discussion With only the holotype and a fragment of body chamber of Reginaites zulu at their disposal, the authors (Klinger & Kennedy 1980a) were initially CRETACEOUS FAUNAS FROM SOUTH AFRICA 145 impressed by the strong ornament on the inner whorls of the phragmocone and the lack of ornament on the body chamber, and failed to recognize the resemblance of the species to Plesiotexanites (P.) stangeri (Baily, 1955). The new material shows that there is remarkable similarity, and that the two species may be closer related than previously suspected. (Young 1963: 93 had also Fig. 23. Reginaites zulu Klinger & Kennedy, 1980. SAM-—PCZ5955, specimen with sparse costation, comparable with Plesiotexanites (P.) stangeri ‘var. sparsicosta’ Spath. From locality 105, St Lucia Formation, Zululand. x 1. 146 ANNALS OF THE SOUTH AFRICAN MUSEUM ->—— = @ 10 a el Fig. 24. Reginaites zulu Klinger & Kennedy, 1980. Part of external suture of SAM-PCZ5956. Millimetre scale bar for size. commented on similarities between Reginaites durhami Young and Plesiotexan- ites (P.) stangeri). In addition, this is the largest assemblage of Reginaites known, most other species being monotypic. The ontogenies of R. zulu and P. (P.) stangeri are very similar, although the accompanying changes in ornament and whorl section are much more strongly expressed in the former. These changes involve development of tuberculation on the inner whorls, gradual effacement of all ornament on the outer phragmocone whorls and body chamber, and change in whorl section. In the very early stages, R. zulu and P. (P.) stangeri differ only in ventral ornament. R. zulu is distinctly tricarinate, whereas P. (P.) stangeri is unicarin- ate, with a pair of spirally elongated external (5) tubercles. In both species the ventrolateral (3 + 4) tubercle, which is situated in the umbilical seam, is enlarged in Protexanites (Protexanites) fashion in early stages of growth. At varying diameters this ventrolateral tubercle starts dividing into two, resulting in a clavate marginal (4) and a generally rounded submarginal (3) tubercle—the so-called ‘Parabevahites’ stage. The Paratexanites-like stage, in which total separation of these tubercles takes place, is completed in both species, generally with the marginal (4) tubercle situated slightly ahead of the CRETACEOUS FAUNAS FROM SOUTH AFRICA 147 Fig. 25. Plesiotexanites (P.) stangeri (Baily, 1855). Cast of holotype, BMNH C73333, from Umzamba estuary, Pondoland. x 0,43. submarginal (3) one. While the separation of the ventrolateral tubercle takes place in P. (P.) stangeri, a lateral (2) tubercle appears, thus completing the pentatuberculate Texanites-like stage. The stage of appearance, strength and persistence of the lateral (2) tubercle in P. (P.) stangeri is very variable. The lateral tubercle may be faint throughout, or persist only for a short time before being absorbed into the ribbing. As far as Jateral ornament is concerned, the only difference between P. (P.) stangeri and R. zulu is the variable appearance 148 ANNALS OF THE SOUTH AFRICAN MUSEUM of a lateral (2) tubercle in the former. The venters, however, remain different throughout. Apart from relative strength of ornament at comparable stages of develop- ment (which may partially be explained by differential preservation), R. zulu and P. (P.) stangeri show comparable intraspecific variation in density of ribbing, which has led to the usage of varietal names sparsicosta and densicosta in the latter species (e.g. Spath 1921: 138, pl. 5 (figs 1-2)). In P. (P.) stangeri these differences in density of ribbing appear to be restricted mainly to the inner whorls, whereas in R. zulu they are noticeable to greater diameters. In both species ornament weakens during ontogeny, although the reduc- tion is more prominent in R. zulu than in P. (P.) stangeri. The new material referable to R. zulu (Fig. 22) shows that parts of the phragmocone may already be devoid of ornament, whereas the body chamber is completely smooth, except for low, broad, undulating swellings (Fig. 21A-—B herein) (Klinger & Kennedy 1980a Fig. 94A-B). In P. (P.) stangeri a similar weakening takes place, but in the holotype the largest specimen as yet obtained of the species, the body chamber (which consists of one whorl here) still bears faint, though discernible lateral ornament (Fig. 25). There are also comparable ontogenetic changes in whorl section (Fig. 17A-B). Both species are initially depressed, quadrate to rectangular, but whorl height gradually increases over whorl breadth, so that the section is eventually distinctly compressed. The venter becomes fastigiate in both species, but in R. zulu this is enhanced by the lateral keels. Phylogenetic relationships between R. zulu and P. (P.) stangeri are not clear, but the ontogenetic changes in both species are too similar to consider them to be entirely those of homoeomorphs. P. (P.) stangeri occurs mainly in the Upper Santonian of. Pondoland, but has also been recorded from the Middle Santonian of Madagascar, and Young (1963: 22) even records it as early as the lowest parts of the Lower Santonian of the Gulf Coast of North America. R. zulu occurs in Zululand in association with Inoceramus (Cordi- ceramus) muelleri Petrascheck, but below the first occurrence of Submortoni- ceras, which places it in the uppermost part of the Santonian. Because of similarities of ornament on the inner whorls of R. zulu and Plesiotexanites (P.) matsumotoi Klinger & Kennedy, 1980a, it seems feasible to derive R. zulu from the latter (probably via Reginaites reymenti Klinger & Kennedy, 1980a) in the Upper Santonian, rather than directly from P. (P.) stangeri. However, the parallel changes in ontogenetic development of R. zulu and P. (P.) stangeri are so striking that we can perhaps trace their ancestry to a common origin somewhere in the Lower Santonian. Palaeobiogeographic data should be interpreted with caution on the basis of this limited material but, to date, the great majority of specimens of P. (P.) stangeri have been reported from the transgressive shallower water sediments of the Umzamba Formation, whereas all known specimens of R. zulu have been found in the deeper water sediments of the St Lucia Formation of Zululand. CRETACEOUS FAUNAS FROM SOUTH AFRICA 149 Differences between R. zulu and other species of Reginaites were given earlier (Klinger & Kennedy 1980a), and are here briefly repeated: R. quadri- tuberculatum Reyment, 1957, R. leei (Reeside, 1927), R. durhami Young, 1963, R. reymenti Klinger & Kennedy, 1980a, and R. gappi Wiedmann, 1978, all have a lateral tubercle at some stage. ‘Reymentites’ hatai Matsumoto, 1965, has the umbilical tubercle displaced on the flanks. All these data support the authors’ (Klinger & Kennedy 1980a) and Young’s (1963) views of placing the genus Reginaites in the subfamily Texaniti- nae, rather than in the subfamily Peroniceratinae. COLOUR-BANDING IN MICROMORPHS OF Submortoniceras woodsi (SPATH, 1921) AND THEIR HOMOEOMORPHY WITH Protexanites (P.) bontanti shimizui Matsumoto, 1970. (By H. C. Klinger & W. J. Kennedy) Two small specimens of Submortoniceras woodsi from the Santonian— Campanian sediments of Kwa-Mbonambi, Zululand (locality 6 of Kennedy & Klinger 1975: 282) are of interest in showing traces of colour bands in the preserved part of the shell, and in apparently being adult at diameters well below those of normal representatives of the species. They also show striking homoeomorphy with the stratigraphically older Protexanites (P.) bontanti shi- mizui Matsumoto, 1970. Submortoniceras woodsi s.\. (Spath, 1921) Figs 26-27 Submortoniceras woodsi s.|. (Spath): Klinger & Kennedy, 1980a: 242, figs 186-199, 200B (with synonomy). Material SAM-PCZ5957-8 both from an unspecified horizon near the Santonian— Campanian boundary at Kwa-Mbonambi (locality 6 of Kennedy & Klinger 1975: 282), St Lucia Formation, Zululand. Dimensions Wb Wh Wb:Wh U PCZ5957 46,0 13,5(34,0) 18,5(40,2) 0,73 15,0(32,6) PCZ5958 48,0 25 (2610) 17,0(35,4) OWS — (crushed) Description Both specimens are less than 50 mm in diameter. PCZ5958 is crushed dorsoventrally, hence the umbilicus appears much narrower than that of 7595/7. 150 ANNALS OF THE SOUTH AFRICAN MUSEUM E F G Fig. 26. Submortoniceras woodsi (Spath, 1921). A-D. SAM-PCZ5957. D without whitening to show parts of colour banding. E-G. SAM-—PCZ5958, specimen crushed dorsoventrally. Both from locality 6, St Lucia Formation, Zululand. All x 1. CRETACEOUS FAUNAS FROM SOUTH AFRICA Si Fig. 27. Submortoniceras woodsi (Spath, 1921). A. Whorl section of SAM-—PCZ5957 to illustrate distribution of colour banding, x 1. B. External suture of SAM—PCZ5956, millimetre scale bar for size. PCZ5957 (Fig. 26A-—D) shows details of the ontogeny. At 5 mm diameter the flanks are still completely smooth; at 8 mm diameter umbilical (1) tubercles and bifurcating ribs are visible on the flanks. At 16 mm diameter weak lateral (2) and submarginal (3) tubercles appear on the flanks. A full texanitine complement of 5 rows of tubercles and a median keel is visible at 25 mm diameter. The lateral (2) and submarginal (3) tubercles, however, are very weak, mimicking a Protexanites (P.)-like type of ornament. Ornament remains like this until near the apertural end, where the lateral (2) tubercle becomes slightly more prominent. The last half-whorl is non-septate and the last few septa overlap. Part of the original aragonitic shell material is preserved. The inner shelly layer is creamy coloured with a reddish tint, whereas the outer layer is more 152 ANNALS OF THE SOUTH AFRICAN MUSEUM brownish. On the outer layer, two darker, sepia-coloured spiral bands are visible, one near the submarginal (3) tubercle (Fig. 26D). In addition, a distinct black substance occurs in the ventral area between the opposite rows of marginal (4) and external (5) tubercles in places. The distribution of colour bands is shown in Figure 27A. PCZ5958 (Fig. 26E-G) is not as well preserved, but shows distinct penta- tuberculate ornament at 25 mm diameter. The lateral (2) and submarginal (3) tubercles are again weaker than the others. Ornament weakens towards the later part of the phragmocone, but again becomes stronger on the body chamber. The last few septa are also crowded. Discussion Because of the weak development of the lateral (2) and submarginal (3) tubercles, and also because of the small size,- these specimens are striking homoeomorphs of the stratigraphically older Protexanites (P.) bontanti shimizui Matsumoto, 1970 (compare Matsumoto 1970, pl. 31 (figs 1-2), text-fig. 6; 1971: 147, pl. 23 (fig 4), text-fig 10; Matsumoto & Hirano 1976: 335, pl. 35 (figs 1-4), text-fig. 2). Close examination, however, clearly shows the pentatuberculate ornament in the Zululand specimens. The two species also differ considerably in details of the suture lines (compare Fig. 27B herein with Matsumoto 1970, text-fig. 6) and cannot be confused with each other. Examination of the suture lines in the present material suggests closest affinity with the subspecies Submortoniceras woodsi woodsi as interpreted by Klinger & Kennedy (1980a: 252-3). Matsumoto (1970: 238) had already commented on the similarity between Protexanites (P.) bontanti s.1. and some examples of Submortoniceras, e.g. S. collignoni (Shimizu); the latter a possible synonym of S. woodsi according to the authors (Klinger & Kennedy 1980a: 238). Matsumoto’s observations are further enhanced by the present data. The presence and distribution of the colour bands on the one specimen, PCZ5957, compares favourably with that described in P. (P.) bontanti shimizui by Matsumoto & Hirano (1976) and adds to the homoeomorphic resemblance. Unfortunately, our material is too poorly preserved to add to Matsumoto & Hirano’s general comments on coloration in Texanitinae, but nevertheless represents yet another record of coloration in the subfamily. The ontogenetic changes in ornament and the crowding of the last few septa suggest that these two specimens represent adult individuals. The ques- tion that now arises is the relationship of these micromorphs to the more commonly known large specimens (even the specimen considered as dwarfed by the authors (Klinger & Kennedy 1980a figs 198, 200B) is gigantic compared to the present material). If nutritional deficiency were to blame (compare Boletzky & Wiedmann 1978), the co-occurrence of ‘normally sized’ specimens, e.g. SAM-PCZ5897 (Klinger & Kennedy 1980a, fig. 192) at 226 mm diameter at the same locality would be a contradiction. Apart from differences in overall CRETACEOUS FAUNAS FROM SOUTH “AFRICA 153 size, the present specimens differ from contemporary S$. woodsi only in having a lower whorl section, as comparison of dimensions shows: D Wb Wh Wo: Wh U U-E ths PCZ5897 226.0 74,0(32.7) 92.0(40,7) 0,80 74,0(32.7) 25:34 PCZ5957 46.0 —13,5(34.0) 17,0(35.4) 0.73 15,0(32.6) 26:36 It is possible that we are here dealing with sexual dimorphs, the ‘normally sized’ specimens being the female (macroconch) and the smaller form the male (microconch). Apart from the fact that this would be the first suggestion of sexual dimorphism in the ammonite subfamily Texanitinae, the implications on the systematics, especially in the genus Submortoniceras, are disturbing. Matsu- moto (1959) and Klinger & Kennedy (1980a) have illustrated the bewildering extent of intraspecific variation in Submortoniceras chicoense, and S. woodsi and §. condamyi respectively. Added to this is a very rapid rate of evolution, as exemplified by S. woodsi and S. condamyi, which makes it difficult to separate one chronospecies satisfactorily from another. Should sexual dimorphism be added to the list of variables in the systematics of the genus Submortoniceras, global correlation may become even more tenuous than at present. ACKNOWLEDGEMENTS We thank Dr F. Stojaspal of the Geologische Bundesanstalt, Vienna; Mag. A. Sikora of the Haus der Natur, Salzburg; Dr H. Kohl of the Ober@sterreich- isches Landesmuseum, Linz; Dr J. Louail of the Faculté des Sciences, Paris; Dr M. K. Howarth and Mr D. Phillips, British Museum (Natural History), London; and Prof. Dr J. Wiedmann, Tubingen, for allowing us to study material in their care. We also thank F. and K. Bohm, private collectors in Salzburg, for allowing us to study their collections. Assistance from the technical staff of the University Museum, Oxford; South African Museum, Cape Town; and Institut und Museum fur Geologie und Palaontologie, Tubingen, is appreciated. Financial aid to H. C. Klinger from the South African Council for Scientific and Industrial Research, and, during the tenure of a research fellow- ship, from the Alexander von Humboldt Foundation, and to W. J. Kennedy from the National Environment Research Council (Great Britain) is gratefully acknowledged. REFERENCES Batty, W. H. 1855. Description of some Cretaceous Fossils from South Affica, collected by Capt. Garden of the 45th Regiment. Q. Jl geol. Soc. Lond. 11: 454-465. Blanckenhorn, M. 1905. Geologie der naheren Umgebung von Jerusalem. Z. dt. Palast. V. 28: 73-120. Boterzxy, S. V. & WiepMANN, J. 1978. Schulp-Wachstum bei Sepia officinalis in Abhangigkeit von Okologischen Parametern. Neues Jb. Geol. Palaont. Abh. 157: 103-106. BRINKMANN, R. 1935. Die Ammoniten der Gosau und des Flysch in den nordostlichen Ostalpen. Mitt. geol. -palaont. Inst. Univ. Hamburg 15: 1-14. 154 ANNALS OF THE SOUTH AFRICAN MUSEUM CoLLiGnNon, M. 1948. Ammonites néocrétacées du Menabe (Madagascar). 1. Les Texanitidae. Annls géol. Serv. Min. Madagascar 13: 7-60 (63-116). CoLLIGNON, M. 1966. Atlas des fossiles caracteristiques de Madagascar (Ammonites) XIV (Santonien). Tananarive: Service Géologique. DIENER, C. 1925. Fossilium catalogus 1: Animalia 29, Ammonoidea neocretacea. J Berlin: Junk. FABRE-TAxy, S. 1963. Faunes du Coniacien et du Santonien de Provence. 1. Les Ammonites du bassin de Beausset (Var.). Annls Paléont. 49: 1-28(99-126). GrossouvrE, A. DE. 1894. Recherches sur la Craie supérieure. II. Paléontologie. Les Ammon- ites de la Craie supérieure. Mém. Carte géol. dét. France 1893: 1-264. Haas, O. 1942. Some Upper Cretaceous ammonites from Angola. Am. Mus. Novit. 1182: 1-24. Hauer, F. Von. 1859. Uber die Cephalopoden der Gosauschichten. Beitr. Geol. Ost. 1: 7-14. KENNEDY, W. J. & KLINGER, H. C. 1975. Cretaceous faunas from Zululand and Natal, South Africa. Introduction, Stratigraphy. Bull. Br. Mus. nat. Hist. (Geol) 25: 265-315. KENNEDY, W. J. & KOLLMANN, H. A. 1977. In: KoLiMANn, H. A. ef al. Beitrage zur Stratigraphie und Sedimentation der Oberkreide des Festlandsockels im ndordlichen Niederosterreich. Jb. geol. Bundesantst. Wien 120: 401-447. KLINGER, H. C. & KENNEDY, W. J. 1980a. Cretaceous faunas from Zululand and Natal, South Africa. The ammonite subfamily Texanitinae Collignon, 1948. Ann. S. Afr. Mus. 80: 1-356. KLINGER, H. C. & KENNEDY, W. J. 1980b. Cretaceous faunas from Zululand and Natal, South Africa. A new sextuberculate texanitid. Ann. S. Afr. Mus. 82: 321-331. KULLMANN, J. & WIEDMANN, J. 1970. Significance of sutures in phylogeny of Ammonoidea. Paleont. Contr. Univ. Kans. 44: 1-32. Matsumoto, T. 1959. Upper Cretaceous Ammonites of California. Il. Mem. Fac. Sci. Kyushu Univ. (D) Special Volume 1: 1-172. Matsumoto, T. 1965. A Monograph of the Collignoniceratidae from Hokkaido, Part 1. Mem. Fac. Sci. Kyushu Univ. (D). 16: 1-80. Matsumoto, T. 1970. A Monograph of the Collignoniceratidae from Hokkaido, Part IV. Mem. Fac. Sci. Kyushu Univ. (D). 20: 225-304. Matsumoto, T. 1971. A Monograph of the Collignoniceratidae from Hokkaido, Part V. Mem. Fac. Sci. Kyushu Univ. (D). 21: 129-162. + Matsumoto, T. 1978. In: Matsumoto, T. & HARAGucHI, Y. A new Texanitine ammonite from Hokkaido. Trans. Proc. palaeont. Soc. Japan (N.S.) 110: 306-318. Matsumoto, T. & Hirano, H. 1976. Colour patterns in some Cretaceous ammonites from Hokkaido. Trans. Proc. palaeont. Soc. Japan (N.S.) 102: 334-342. Matsumoto, T. & UsepA, U. 1962. In Ueda, Y. The type Himenoura Group. Mem. Fac. Sci. Kyushu Univ. (D). 12: 129-178. = PERVINQUIERE, L. 1907. Etudes de paléontologie tunisienne. 1. Céphalopodes des terrains secondaires. Mém. Carte géol. Tunis 1907: 1-438. Pop, G. & Szasz, L. 1973. Santonianul diu reginunea hategului (Carpatii Meridionali). Studii Cerc. Geol. 18: 463-467. REDTENBACHER, A. 1873. Die Cephalopodenfauna der Gosauschichten in den nordéstlichen Alpen. Abh. Geol. Bundesanst. Wien 5: 91-140. REESIDE, J. B. 1927. Cephalopods of the Eagle Sandstone and related Formations in the Western Interior of the United States. Prof. Pap. U.S. geol. Surv. 151: 1-87. REYMENT, R. A. 1957. Uber einige wirbellose Fossilien aus Nigerien und Kamerun, West- Afrika. Palaeontographica 109: 41-70. Roemer, F. A. 1852. Die Kreidebildungen von Texas und ihre organischen Einschliisse. Bonn: Adolph Marcus. SCHLUTER, C. 1871-76. Die Cephalopoden der oberen deutschen Kreide. Palaeontographica 21: 1-24 (1871), 25-120 (1872), 24: 121-262 (1876). SpATH, L. F. 1921. On Cretaceous Cephalopoda from Zululand. Ann. S. Afr. Mus. 12: 217-321. SpaTH, L. F. 1922. On the Senonian ammonite fauna of Pondoland. Trans. R. Soc. S. Afr. 10: 113-147. THOMEL, G. 1969. Sur quelques ammonites Turoniennes et Sénoniennes nouvelles ou peu connues. Annls Paléont. 55: 1-11 (108-119). CRETACEOUS FAUNAS FROM SOUTH AFRICA 155 WEDEKIND, R. 1916. Uber Lobus, Suturallobus und Inzision. Zentbl. Miner. Geol. Paldont. 1916: 185-195. WIEDMANN, J. 1978. Eine palaogeographisch interessante Ammonitenfaunula aus der alpinen Gosau (Santon, Becken von Gosau, Oberésterreich). Eclog. geol. Helv. 71/3: 663-765. WIEDMANN, J. 1979 Jn: HERM, D., KAUFFMAN, E. G. & WIEDMANN, J. The age and depositional environment of the “Gosau’’-group (Coniacian-Santonian), Brandenberg/Tirol, Austria. Mitt. bayer. St. Paldont. Hist. Geol. 19: 27-92. YaBE, H. & Suimizu, H. 1923. A note on the genus Mortoniceras. Jap. J. Geol. Geogr. 2: 27-30. Younc, K. 1963. Upper Cretaceous ammonites from the Gulf Coast of the United States. Univ. Tex. Publs 6304: 1-373. ZuRcHER, P. 1905. Communication. Bull. Soc. géol. Fr. 5: 686. wr i t } As 6. SYSTEMATIC papers must conform to the International code of zoological nomenclature (particularly Articles 22 and 51). Names of new taxa, combinations, synonyms, etc., when used for the first time, must be followed by the appropriate Latin (not English) abbreviation, e.g. gen. nov., sp. nov., comb. nov., syn. nov., etc. An author’s name when cited must follow the name of the taxon without intervening punctuation and not be abbreviated; if the year is added, a comma must separate author’s name and year. The author’s name (and date, if cited) must be placed in parentheses if a species or subspecies is transferred from its original genus. The name of a subsequent user of a scientific name must be separated from the scientific name by a colon. Synonymy arrangement should be according to chronology of names, i.e. all published scientific names by which the species previously has been designated are listed in chronological order, with all references to that name following in chronological order, e.g.: Family Nuculanidae Nuculana (Lembulus) bicuspidata (Gould, 1845) Figs 14-15A Nucula (Leda) bicuspidata Gould, 1845: 37. Leda plicifera A. Adams, 1856: 50. Laeda bicuspidata Hanley, 1859: 118, pl. 228 (fig. 73). Sowerby, 1871: pl. 2 (fig. 8a—b). Nucula largillierti Philippi, 1861: 87. Leda bicuspidata: Nicklés, 1950: 163, fig. 301; 1955: 110. Barnard, 1964: 234, figs 8-9. Note punctuation in the above example: comma separates author’s name and year “semicolon separates more than one reference by the same author full stop separates references by different authors figures of plates are enclosed in parentheses to distinguish them from text-figures dash, not comma, separates consecutive numbers Synonymy arrangement according to chronology of bibliographic references, whereby the year is placed in front of each entry, and the synonym repeated in full for each entry, is not acceptable. In describing new species, one specimen must be designated as the holotype; other speci- mens mentioned in the original description are to be designated paratypes; additional material not regarded as paratypes should be listed separately. The complete data (registration number, depository, description of specimen, locality, collector, date) of the holotype and paratypes must be recorded, e.g.: Holotype SAM-—A13535 in the South African Museum, Cape Town. Adult female from mid- tide region, King’s Beach Port Elizabeth (33°51’S 25°39’E), collected by A. ‘Smith, 15 January 1973. Note standard form of writing South African Museum registration numbers and date. 7. SPECIAL HOUSE RULES Capital initial letters (a) The Figures, Maps and Tables of the paper when referred to in the text e.g. *... the Figure depicting C. namacolus ...’; *. .. in C. namacolus (Fig. 10)...’ -(b) The prefixes of prefixed surnames in all languages, when used in the text, if not preceded by initials or full names e.g. Du Toit but A.L.du Toit; Von Huene but F. von Huene (c) Scientific names, but not their vernacular derivatives e.g. Therocephalia, but therocephalian Punctuation should be loose, omitting all not strictly necessary Reference to the author should be expressed in the third person Roman’ numerals should be converted to arabic, except when forming part of the title of a book or article, such as ‘Revision of the Crustacea. Part VIII. The Amphipoda.’ Specific name must not stand alone, but be preceded by the generic name or its abbreviation to initial capital letter, provided the same generic name is used consecutively. Name of new genus or species is not to be included in the title: it should be included in the abstract, counter to Recommendation 23 of the Code, to meet the requirements of Biological Abstracts. WILLIAM JAMES KENNEDY HERBERT CHRISTIAN KLINGER & HERBERT SUMMESBERGER CRETACEOUS FAUNAS FROM ZULULAND AND NATAL, SOUTH AFRICA ADDITIONAL OBSERVATIONS ON THE AMMONITE SUBFAMILY TEXANITINAE COLLIGNON, 1948 OCTOBER 1981 | ISSN 0303-2515 CAPE TOWN INSTRUCTIONS TO AUTHORS 1. MATERIAL should be original and not published elsewhere, in whole or in part. 2. LAYOUT should be as follows: (a) Centred masthead to consist of Title: informative but concise, without abbreviations and not including the names of new genera or species Author’s(s’) name(s) Address(es) of author(s) (institution where work was carried out) Number of illustrations (figures, enumerated maps and tables, in this order) (b) Abstract of not more than 200 words, intelligible to the reader without reference to the text (c) Table of contents giving hierarchy of headings and subheadings (d) Introduction (e) Subject-matter of the paper, divided into sections to correspond with those given in table of contents (f) Summary, if paper is lengthy (g) Acknowledgements (h) References (Gi) Abbreviations, where these are numerous 3. MANUSCRIPT, to be submitted in triplicate, should be typewritten and neat, double spaced with 2,5 cm margins all round. First lines of paragraphs should be indented. Tables and a list of legends for illustrations should be typed separately, their positions indicated in the text. All pages should be numbered consecutively. Major headings of the paper are centred capitals; first subheadings are shouldered small capitals; second subheadings are shouldered italics; third subheadings are indented, shouldered italics. Further subdivisions should be avoided, as also enumeration (never roman numerals) of headings and abbreviations. Footnotes should be avoided unless they are short and essential. Only generic and specific names should be underlined to indicate italics; all other marking up should be left to editor and publisher. 4. ILLUSTRATIONS should be reducible to a size not exceeding 12 « 18 cm (19 cm including legend); the reduction or enlargement required should be indicated; originals larger than 35 x 47 cm should not be submitted; photographs should be rectangular in shape and final size. A metric scale should appear with all illustrations, otherwise magnification or reduction should be given in the legend; if the latter, then the final reduction or enlargement should be taken into consideration. All illustrations, whether line drawings or photographs, should be termed figures (plates are not printed; half-tones will appear in their proper place in the text) and numbered in a single series. Items of composite figures should be designated by capital letters; lettering of figures is not set in type and should be in lower-case letters. The number of the figure should be lightly marked in pencil on the back of each illustration. 5. REFERENCES cited in text and synonymies should all be included in the list at the end of the paper, using the Harvard System (ibid., idem, loc. cit., op. cit. are not acceptable): (a) Author’s name and year of publication given in text, e.g.: ‘Smith (1969) describes .. .” ‘Smith (1969: 36, fig. 16) describes...’ ‘As described (Smith 1969a, 1969b; Jones 1971)’ ‘As described (Haughton & Broom 1927)...’ ‘As described (Haughton et al. 1927)...’ Note: no comma separating name and year Dagination indicated by colon, not p. names of joint authors connected by ampersand et al. in text for more than two joint authors, but names of all authors given in list of references. (b) Full references at the end of the paper, arranged alphabetically by names, chronologically within each name, with suffixes a, b, etc. to the year for more than one paper by the same author in that year, e.g. Smith (1969a, 19695) and not Smith (1969, 1969a). For books give title in italics, edition, volume number, place of publication, publisher. For journal article give title of article, title of journal in italics (abbreviated according to the World list o, scientific periodicals. 4th ed. London: Butterworths, 1963), series in parentheses, volume number, part number (only if independently paged) in parentheses, pagination (first and last pages of article). Examples (note capitalization and punctuation) BULLOUGH, W. S. 1960. Practical invertebrate anatomy. 2nd ed. London: Macmillan. FISCHER, P.—H. 1948. Données sur la résistance et de le vitalité des mollusques. J. Conch., Paris 88: 100-140. FiscHER, P.-H., DuvaL, M. & Rarry, A. 1933. Etudes sur les échanges respiratoires des littorines. Archs Zool. exp. gén. 74: 627-634. \ Konn, A. J. 1960a. Ecological notes on Conus (Mollusca: Gastropoda) in the Trincomalee region of Ceylon. Ann. Mag. nat. Hist. (13) 2: 309-320. Konn, A. J. 19606. Spawning behaviour, egg masses and larval development in Conus from the Indian Ocean. Bull. Bingham oceanogr. Coll. 17 (4): 1-51. THEELE, J. 1910. Mollusca: B. Polyplacophora, Gastropoda marina, Bivalvia. In: SCHULTZE, L. Zoologische und anthropologische Ergebnisse einer Forschungsreise im westlichen und zentralen Siid-Afrika 4: 269-270. Jena: Fischer. Denkschr. med.-naturw. Ges. Jena 16: 269-270. (continued inside back cover) ANNALS OF THE SOUTH AFRICAN MUSEUM ANNALE VAN DIE SUID-AFRIKAANSE MUSEUM Volume 86 Band October 1981 Oktober Part 5 Deel cee TRESS S VVVIWT SS S OCCLUSAL MORPHOLOGY OF THE MANDIBULAR PERMANENT MOLARS OF THE SOUTH AFRICAN NEGRO AND THE KALAHARI SAN (BUSHMAN) By EG RENE Cape Town Kaapstad The ANNALS OF THE SOUTH AFRICAN MUSEUM are issued in parts at irregular intervals as material becomes available Obtainable from the South African Museum, P.O. Box 61, Cape Town 8000 Die ANNALE VAN DIE SUID-AFRIKAANSE MUSEUM word uitgegee in dele op ongereelde tye na gelang van die beskikbaarheid van stof Verkrygbaar van die Suid-Afrikaanse Museum, Posbus 61, Kaapstad 8000 OUT OF PRINT/UIT DRUK iy DCE}, S39), 30D, ESS Stn), SEBS, TD), BCG =o), LO SCE WOES) WGD, 5 Tico), SCS), 2), 2, G3), 226), 33, 450) EDITOR/REDAKTRISE Ione Rudner Copyright enquiries to the South African Museum Kopieregnavrae aan die Suid-Afrikaanse Museum ISBN 0 86813 023 0 Printed in South Africa by In Suid-Afrika gedruk deur The Rustica Press, Pty., Ltd., Die Rustica-pers, Edms., Bpk., Court Road, Wynberg, Cape Courtweg, Wynberg, Kaap OCCLUSAL MORPHOLOGY OF THE MANDIBULAR PERMANENT MOLARS OF THE SOUTH AFRICAN NEGRO AND THE KALAHARI SAN (BUSHMAN) By F, E. GRINE South African Museum, Cape Town (With 8 figures and 26 tables) [MS accepted 2 July 1981] ABSTRACT A number of morphological crown variants are investigated. These features include the principal cusp number, the tuberculum sextum and tuberculum intermedium, the configura- tion of the primary occlusal pattern, the mesial and distal foveae, the deflecting wrinkle, and the appearance of the occlusal enamel. The definitions and the various methods employed in the classification of these characters are reviewed. The amount and degree of significant sexual dimorphism of these crown variants are low in both the San and Negro. Significant interpopulation differences in these features are more common. The fovea anterior and tuberculum sextum appear to be the most useful discriminatory characters, while the second molar displays the greatest number of significant differences between the San and Negro. A characterization of the lower molar morphology of the San and Negro is provided. Paired chi-square evaluations of various characters showed most to be independent of one another. The dental data provided here for the San and the South African Negro are compatible with, and to some degree corroborate what is known about, the interrelationships of these peoples from non-dental morphological and genetic data. CONTENTS PAGE Introduction 4 : : : : : ; 5 oo lS8 Material and methods ; : ~~ SS Definition and classification of morphological features ge OPA Principal cusp number . : , ; ; : é Pe G2 Tuberculum sextum . : : : ; : : 5 AOS Tuberculum intermedium . ; ; ; é ‘ . 163 Primary occlusal pattern. j : : : . 164 Deflecting wrinkle . : ; F : : : : . 168 Fovea anterior and posterior : : 5 ; : sof 170 Enamel wrinkling . ; : : : ; : : 5 LIZ Results : , : A ; : 5 3 : : 5 ke Discussion . : : : P : ‘ 5 : 5 NEY Sexual dimorphism : : ; 5 . 199 Characteristics of San and Negro molars ‘ : : . 200 Occlusal morphological interrelationships : 202 Biological interrelationships of the South African Negro and the Kalahari San . ; : : ; : 4.) 203 Summary . : ‘ A 5 ; ; ; ; 5 730 Acknowledgements ; : : ‘ , : : 5 . 208 References . : : ; : 5 : : ‘ : = 209 157 Ann. S. Afr. Mus. 86 (5), 1981: 157-215, 8 figs, 26 tables. 158 ANNALS OF THE SOUTH AFRICAN MUSEUM INTRODUCTION The human mandibular molar presents a number of features that are of importance in both anthropological and palaeoanthropological contexts. These occlusal characters have been recorded extensively in both recent and fossil representatives of hominid populations, and they have been used to quantify both interracial and population differences as well as evolutionary trends. Amongst the numerous minor morphological variants that can be examined, several have been shown to have considerable theoretical potential as anthro- pological markers. These features are: (i) the principal cusp number, (ii) the presence or absence of the tuberculum sextum, (iii) the presence or absence of the tuberculum intermedium, (iv) the form of the primary occlusal pattern, -(v) the presence or absence of a deflecting wrinkle, (vi) the presence or absence of the mesial and distal foveae, and (vii) the appearance of the enamel (smooth or crenulate). The purpose of this paper is to provide information on the morphology of the lower permanent molars of two populations, the South African Negro and the Kalahari San (Bushman), for which the majority of these characters have heretofore not been recorded. Previous investigations of these populations have been restricted to the detailing of cusp number (Shaw 1927; Drennan 1929; Oranje 1934) or of the primary cusp and groove pattern (Kiernberger 1955; Van Reenen 1966; Jacobson 1967). Morris (1970) presented data on the fre- quency of the metaconid deflecting wrinkle on the first molars only. Although these characters have been described and documented exten- sively in the literature, there remains a fundamental lack of consensus amongst workers on the classification and interpretation of several of these features. This lack of agreement presents one of the major obstacles which preclude comparisons between results reported in the various studies of these features in recent human populations. Thus, in addition to providing information on the morphology of the South African Negro and San lower molar, the problems of definition and classification of these characters will also be considered in the hope that a workable series of models might be developed. MATERIAL AND METHODS The dentitions examined are drawn from the Khoisanoid and Negroid populations of southern Africa. The Negroid sample comprises skulls of South African Bantu-speaking Negroes housed in the Raymond A. Dart Collection of Human Skeletons, Department of Anatomy, University of the Witwaters- rand. The skeletal material housed in this collection has been prepared from dissection-hall subjects and, thus, the sex of each individual is known and the tribal group is known for most. The Negro mandibles examined here were drawn from the following tribal groups: Natal Nguni (Zulu and Swazi), Cape Neguni (Xhosa and Pondo), Sotho (Sotho and Tswana), and Shangana—Tonga PERMANENT MOLARS OF NEGRO AND SAN 159 (Shangaan and Tonga). The major group only (e.g. Natal Nguni, Sotho) was recorded for each individual because De Villiers (1968) has shown through analyses of skull variability that there are no distinguishing intertribal differences. Although some of the material housed in the Raymond A. Dart Collection is probably derived from individuals who lived on the Witwatersrand for perhaps their entire lives, or at least prior to their deaths, some (or their parents) had moved to the Witwatersrand from their ‘traditional’ tribal areas. The geographical distribution of the tribal groups from which material was selected for study is shown in Figure 1. The numbers of individuals derived from each group are recorded in Table 1. A total of 506 Negro dentitions (353 male and 153 female) was examined. The term San, rather than Bushman, is used in this study as the former is taken to denote a biological entity whereas the latter refers more precisely SOUTH WEST AFRICA NAMIBIA BOTSWANA HaSsonacea eecce RBeooGanod eeeecee; . eee SoS eecccce Wsccasosaee Nooes6e6ac ecco eecce eeocccce eccccor eocee oP s000ce Areeccccece eecccccce eee ®@-eece eececce eececeo e aoe oe ~#aaeanegee S22GS a Ps “7 > SOUTH AFRICA oop Oks mefls - & Nan ae SSS CWA TH, Fig. 1. Distribution of the South African Negro population groups sampled in the present study. 1—Natal Nguni, 2.—Cape Nguni, 3—Sotho, 4—Shangana—Tonga (adapted from Lawton 1967). 160 ANNALS OF THE SOUTH AFRICAN MUSEUM to the linguistic and ethnological practices of these people (Jenkins & Tobias 1977; Tobias 1978). TABLE 1 Number of Negro and San individuals examined. Population Tribal or linguistic group Male Female Total Negro Natal Nguni : : ; ; : : 111 60 171 Cape Nguni . 3 : ; ‘ ; : 90 34 124 SOMO) a wee ec ke Dean 99 55 154 Venda . : ; 3 : : ; , Dil 3 24 Shangana—Tonga : ‘ 3 ; : 32 y 34 353 154 507 San Nharo . : : : P ‘ ; : 16 14 30 G/wi_. Ree ey Am san |” alas 6 3 9 ‘Tshu-Khwe’ cee eke gaa emma 8 y 10 kx ?ao// %€T : : : 5 : : 5 5 10 IXOOtr tee et Pa ee ee IS) 2) 36 Unspecified . : ; : : k ; 66 64 130 116 109 7 DI5) The San sample is comprised of plaster casts made from impressions of living subjects taken by Van Reenen and his co-workers on three separate trips to the Kalahari in 1958, 1959, and 1964 (Van Reenen 1966). The sex of each individual is known, and the ‘tribal’ group was recorded for a number of the subjects. The San populations recorded are drawn from the Northern, Central and Southern Bush linguistic groups. The Nharo and G/wi form a homogeneous linguistic group which Bleek (1927) termed the Central Bush language group. — Westphal (1963) renamed this unit the Tshu-Khwe linguistic family. Some ten individuals (eight males and two females) were recorded in the field notes as ‘Tshu-Khwe’, and although this is not properly a ‘tribal group’ it is indicated in Table | as a population. The #kx%ao//€i belong to the Northern Bush (Bleek 1927) or Zhu (Westphal 1963) language group, and the !x66 are con- sidered to belong to the Southern Bush language group.* The numbers of individuals recorded for the various ‘tribal’ or linguistic groups are presented in Table |. This information was not available for most of the material examined. A total of 225 casts of San dentitions (116 males and 109 females) was examined. The San casts were made at a number of different localities in Botswana. These places are located, for the most part, over the western side of that country (Fig. 2), and they comprised private farms, small settlements dominated by a trading store, and areas around boreholes. Van Reenen (1964) lists some thirteen different places at which San dentitions were examined. * Thanks to A. Traill (Dept. of Linguistics, University of the Witwatersrand) for these spellings — Editor. PERMANENT MOLARS OF NEGRO AND SAN 161 For each individual only a single tooth, usually the left, of any one tooth type (i.e. M,, M. or Mg) was examined. Thus, throughout, where reference is made to number (N) this refers to both number of teeth and number of indi- ZIMBABWE SOUTH WEST AFRICA/ NAMIBIA ees REPUBLIC ; DEKAR | OF GHANZI ° SUNNYSIDE ne BOTSWANA » © TAKATSHWANE \. @ LONE TREE SOUTH AFRICA Fig. 2. Areas from which most of the San, whose dental casts were examined, were derived. viduals. In each instance only those teeth that were undamaged and not obviously diseased were examined. Berry (1976) has shown that most minor dental variants are destroyed by wear, and, therefore, only those teeth that had suffered minimal abrasion or attrition were included. Not every individual examined presented a full dental complement. Data were recorded on coded sheets and then transferred to computer punch cards. The data were then processed and ‘cleaned’ via the OSIRIS programs package, and analysed via the SPSS programs package (Nie ef al. 1975). 162 ANNALS OF THE SOUTH AFRICAN MUSEUM DEFINITION AND CLASSIFICATION OF MORPHOLOGICAL FEATURES PRINCIPAL CUSP NUMBER The protoconid, metaconid, hypoconid, entoconid, and hypoconulid are the principal cusps. The criterion accepted here for the presence of a cusp was that it represented a distinct entity, demarcated by deep grooves from adjacent cusps. The number of cusps present were recorded. No attempt was made to quantify the sizes of the cusps. A survey of the literature indicated this mode of notation to be the most widely used; a much smaller number of authors have recorded ‘half-sized’ cusps (Drennan 1929; Snyder et al. 1969; Corruccini 1972), and an even smaller number have utilized either Hrdlicka’s (1910) or Janzer’s (1927) notation schemes for scoring reduced cuspal size (Goldstein 1931; Kiernberger 1955). Neither Hrdli¢ka’s nor Janzer’s method of counting Fig. 3. Drawing of a generalized mandibular left molar showing some of the features examined in this study. e—entoconid, h—hypoconid, hl—hypoconulid, m—metaconid, p—protoconid, pm—postmetaconulid, ti—tuberculum intermedium, ts—tuberculum sextum. PERMANENT MOLARS OF NEGRO AND SAN 163 cusps differentiates between the principal cusps and the secondary cuspulids (e.g. tuberculum sextum and intermedium). Thus, data presented by authors who followed either of these two systems are not directly comparable with the results of studies in which secondary cuspulids have been recorded independently of the principal cusp number. TUBERCULUM SEXTUM This is an occasional cuspulid situated between the entoconid and the hypo- conulid (Fig. 3). Selenka (1898) first proposed the name tuberculum accessorium posteriore internum for such a structure. Hellman (1928) proposed that it be referred to as the C6, and Bennejeant (1936) named it the tuberculum sextum. Matsuda (1961), Turner & Hanihara (1977), Turner & Swindler (1978), and Turner et al. (1969) in the literature accompanying their ‘Cusp 6’ Plaque, have referred to this cuspulid as the entoconulid. However, the entoconulid, or pre- entoconulid of MacIntyre (1966) is located on the entocristid, which is mesial of the tip of the entoconid (Van Valen 1966; Szalay 1969; Hershkovitz 1971), whereas the tuberculum sextum is situated distal to the entoconid. Thus, the tuberculum sextum is the homologue of the postentoconulid of the mammalian molar (Hershkovitz 1971). In some studies the size of this cuspulid relative to the size of the hypo- conulid has been recorded (e.g. Turner 1970, 1976, 1979; Turner & Hanihara 1977; Turner & Swindler 1978). In the present analysis the absence or presence only of the tuberculum sextum was recorded, and its size gradients were not considered. This feature was regarded as present only if a separate cuspal tip was discernible, bounded by distinct grooves separating it from both the ento- conid and hypoconulid. Thus, the tooth classified as category 1 on Turner et al. (1969) ‘Cusp 6’ Plaque would have been regarded here as showing absence of this cuspulid. TUBERCULUM INTERMEDIUM This occasional cuspulid is located between the metaconid and the ento- conid (Fig. 3). Selenka (1898) recorded the presence of this feature on some pongid lower molars and proposed the name tuberculum accessorium mediale internum for it. Schwalbe (1916) and Hellman (1928) referred to this structure respectively as the tuberculum intermedium and the C7. These latter two names are the most frequently encountered in the anthropological literature. Robin- son (1956), in his description of the South African australopithecine teeth, referred to this tubercle as the lingual or median lingual accessory cusp. As pointed out by Jorgensen (1956), it is difficult to ascertain whether various authors who have reported data on this cuspulid have referred, in all cases, to the same structure. The main reason for this ambiguity appears to be the presence in some teeth of two distinct cuspulids between the metaconid 164 ANNALS OF THE SOUTH AFRICAN MUSEUM and entoconid. In such teeth the problem is to determine which of these struc- tures is homologous with the tuberculum intermedium. The cuspulid that lies directly distal to the tip of the metaconid (Fig. 3) has been referred to as the metastylid (Osborn 1888; MacIntyre 1966; Van Valen 1966; Szalay 1969), the tuberculum internum accessorius anterius (Remane 1960), the postmetaconulid (Hershkovitz 1971), and the metaconulid (Turner 1970, 1979; Turner & Hanihara 1977; Turner & Swindler 1978; Turner et al. 1970). The cuspulid that is situated directly mesial to the tip of the entoconid has been called the entostylid (Osborn 1888), the tuberculum internum acces- sorium posterius (Remane 1960), the pre-entoconulid (MacIntyre 1966) and the entoconulid (Van Valen 1966; Szalay 1969; Hershkovitz 1971). Selenka (1898), in his description of the tuberculum accessorium mediale internum, states that it is usually separated from the metaconid by only a shallow groove, it seldom appears as a part of the entoconid, and that in excep- tional cases it shows no continuity with either the metaconid or entoconid. Most workers have stated either explicitly, or indicated in their illustrations of the tuberculum intermedium, that they consider this cuspulid to be homo- logous with the postmetaconulid (Bennejeant 1936; Jorgensen 1956; Robinson 1956; Hanihara 1961; Hanihara & Minamidate 1965; Hershkovitz 1971; Matsuda 1961; Turner & Hanihara 1977; Suzuki & Sakai 1956c, 1973). How- ever, Turner et al. (1970) depict an entoconulid (more precisely a swollen mesial entoconid crest) as a Type 1A development of the tuberculum intermedium in their ‘Cusp 7’ Plaque. In a previous study (Grine 1978), a number of teeth were found to possess both a well-defined tuberculum intermedium and a distinct postmetaconulid. In addition, this study revealed a number of molars in which both the mesial crest of the entoconid and the distal aspect of the metaconid were enlarged and separated from the tips of the cusps. Because of this the author cannot subscribe with full confidence to the homology of the tuberculum intermedium with the postmetaconulid. In the present study the presence of a tuberculum intermedium was recog- nized only where a distinct cuspulid was present between, and well-separated from, both the metaconid and entoconid. Thus, the teeth displayed in Turner et al. (1969) ‘Cusp 7’ Plaque categories 2, 3 and 4 and in Hanihara’s (1961) Plaque D9 categories 2 and 3 are recognized as possessing a tuberculum inter- medium. No attempt was made to quantify the size of this cuspulid. PRIMARY OCCLUSAL PATTERN The Dryopothecus pattern of the lower molar was first defined and described by Gregory (1916). The concept and diagnosis of this pattern were expanded upon by Gregory (1926a, 1934) and Gregory & Hellman (19265). The definition of the pattern eventually included the presence of (i) five principal cusps, (ii) a PERMANENT MOLARS OF NEGRO AND SAN 165 symmetrical arrangement of the grooves surrounding the hypoconid forming the tines and the groove between the metaconid and entoconid forming the stem of a Y, and (ili) a broad basal contact of the metaconid and hypoconid. Gregory (1926a, 19265) noted that in man the symmetrical arrangement characteristic of the Dryopithecus molar has been replaced by a ‘cruciform’ or ‘+ pattern’ where ‘with the crowding forward of the entoconid, the transverse sulcus in front of the entoconid has been brought nearly in line with the trans- verse sulcus between the protoconid and hypoconid’ (1962a: 423). Gregory & Hellman (1926a, 1926b) added that the metaconid—hypoconid contact of the Dryopithecus pattern had been lost in the human molar and had been replaced by a protoconid—entoconid contact, which they considered to be a feature of the ‘cruciform’ or ‘+ pattern’. Hellman (1928) presented a formal scoring scheme and set of definitions by which the occlusal pattern of the lower molar could be classified. He recog- nized four distinct types: Y5, +5, (4 (or Y4), and +4 (Fig. 4). Reference to Figure 4 indicates that a + pattern was considered by him to possess either a protoconid-entoconid contact (Fig. 4C) or a true cruciform point contact (Fig. 4D). Hellman’s (1928) occlusal pattern types were determined according to the arrangement of the lingual and mesiobuccal grooves rather than by the type of cuspal contact exhibited by a tooth. A number of workers have utilized this system of classification. Weidenreich (1937) considered that the cruciform (true point contact of the metaconid, protoconid, hypoconid and entoconid) and _ protoconid— entoconid contacts should be regarded as two distinct types. He proposed the Fig. 4. Left lower molars showing Hellman’s criteria for the classification of the primary occlusal pattern (after Hellman 1928: 165, fig. 7). 166 ANNALS OF THE SOUTH AFRICAN MUSEUM name ‘reversed Dryopithecus pattern’ for the protoconid—entoconid contact (Weidenreich 1937: 96). However, Weidenreich’s suggestion remained unheeded until Senyurek (19525), in a study of the ancient inhabitants of Alaca Héyiik, noticed that several lower molars in his sample exhibited a protoconid—ento- conid contact. He also regarded this as a significant deviation from both the cruciform and Dryopithecus patterns, and he proposed to call this ‘deviant plan .. . the protoconid—entoconid connection’ (19526: 184). It was not until 1955, though, when Jorgensen described the protoconid—entoconid contact as an ‘X pattern’, that the differentiation of Hellman’s + pattern into two distinct morphological entities became widely accepted by students of human dentition. Jorgensen (1955) recognized three types of lower molar pattern: the Y pattern, characterized by a metaconid—hypoconid contact; the + pattern, defined by a point contact of the protoconid, metaconid, hypoconid and entoconid; and the X pattern (Fig. 5). Thus, whereas Hellman’s (1928) scheme employed the Mesial Mesial Alesial Fig. 5. Right lower molars showing Jorgensen’s criteria for the classification of the primary occlusal pattern. A. Y pattern. B. + pattern. C. X pattern. These standards were used in the present study (after Jorgensen 1955: 197, fig. 1). PERMANENT MOLARS OF NEGRO AND SAN 167 Fig. 6. Left mandibular third molars showing the H pattern. A. H pattern with metaconid, protoconid and hypoconid contacting at a point. B. H pattern with metaconid and hypo- conid separated by the protoconid. relationships of the mesiobuccal and lingual grooves, Jorgensen’s (1955) molar types were defined on the basis of cuspal contact. Variants other than Jorgensen’s (1955) three categories are encountered. Senyurek (1952a), for example, noted a single molar (M;) amongst the Chalco- lithic and Copper age dentitions from Anatolia in which the protoconid crest was so well developed as to effect contact with the base of the hypoconulid. 168 ANNALS OF THE SOUTH AFRICAN MUSEUM The present author has found another interesting variant of cuspal contact in some South African Negro and San lower molars (Grine 1978). In this variant the entoconid is prevented by a grossly enlarged tuberculum inter- medium from making contact with either the protoconid or metaconid (Fig. 6). The presence of an enlarged tuberculum intermedium does not necessarily affect the relationships of the protoconid, metaconid and hypoconid, in which case a Y pattern is expressed. However, teeth in which these three cusps contact at a point (Fig. 6A), or in which the metaconid and hypoconid are separated by the protoconid (Fig. 6B), cannot be classified as possessing a conventional + or X pattern. A new category, designated by the letter ‘H’, was proposed for the classification of crowns exhibiting these latter variants (Grine 1978). In the present study the cuspal relationships of molars were classified as Y, +, X (Jorgensen 1955), or H. It must be stressed that the Y pattern, as used here, is not synonymous with the Dryopithecus pattern. Also, the principal cusp number and the primary occlusal pattern were scored separately. Studies of these two characters have shown them to be unrelated (Jorgensen 1955; Garn et al. 1966a, 1966b; Devoto & Cigliano 1970; Devoto et al. 1970) or only very slightly related within individual teeth (Rosen- zweig & Zilberman 1969). DEFLECTING WRINKLE Weidenreich (1937) noted that the dispositions of the crests of the principal cusps may affect the expression of the primary occlusal pattern. He pointed out that on the RM, of the Heidelberg (Mauer) mandible while the mesio- buccal and lingual grooves form a straight transverse fissure, ‘the metaconid is in contact with the hypoconid by means of a special wrinkle crossing the intersection of the furrows’ (Weidenreich 1937: 90). In his description of a LM, of Gigantopithecus blacki, Weidenreich noted a similar structure where ‘the metaconid has a large, well-defined median wrinkle (dw, Plate 10) which emerges from the tip, extends in the direction of the protoconid, and there forms a right angle distally. It transverses the groove between the protoconid and hypoconid and meets the mesial wrinkle of the entoconid approximately in the centre of the occlusal surface . . .” (Weidenreich 1945: 71-72). (Weidenreich’s (1945) plate 10 is reproduced here in Figure 7.) He (1945: 84) proposed the name ‘deflecting wrinkle’ for this structure. A number of studies have been made of this feature in modern human populations (Hanihara 1956, 1963, 1966, 1970; Hanihara et al. 1964, 1975; Suzuki & Sakai 19566; Morris 1970; Turner 1976, 1979; Turner & Hanihara 1977: Turner & Swindler 1978; Axelsson & Kirveskari 1977). Perhaps even more than any other occlusal feature, the deflecting wrinkle has been subjected to widely disparate schemes of classification. Hanihara (1961, Plaque D 10) presented a model for the classification of the deflecting wrinkle in which two types were recognized: (i) the central meta- PERMANENT MOLARS OF NEGRO AND SAN 169 Fig. 7. Left lower molar of Gigantopithecus blacki showing the deflected metaconid crest (dw) which Weidenreich named the deflecting wrinkle (after Weidenreich 1945, pl. 10 (fig. b1)). conid crest was ‘very well developed in its thickness’ (shown by the dm, of his Plaque) and (ii) as shown by the M, of his Plaque, the metaconid crest exhibited a true deflecting wrinkle (true in the sense that this feature was defined by Weidenreich). Hanihara (1961: 42) stated that ‘the well-developed central ridge seems to be homologous with one which is described as a deflecting wrinkle by Weidenreich’. The crests are, in themselves, homologous, but a straight, well-developed crest is not a deflecting wrinkle. Following Hanihara’s mistaken identification of a straight metaconid crest as a deflecting wrinkle, a number of workers (e.g. Hanihara et al. 1975; Morris 1970; Turner 1976, 1979; Turner & Swindler 1978; Axelsson & Kirveskari 1977) have either developed or followed classifications in which a thick, straight metaconid ridge is erroneously recognized as a manifestation of a deflecting wrinkle. 170 ANNALS OF THE SOUTH AFRICAN MUSEUM An essential feature of the deflecting wrinkle, as defined originally by Weidenreich (1945), is that the metaconid crest courses towards the proto- conid and is deflected distally in mid-course. Although Weidenreich described this feature as traversing the longitudinal groove between the hypoconid and entoconid, a deflected crest may be foreshortened so that it does not emerge from between the protoconid and metaconid. Therefore, it is proposed that the deflecting wrinkle be defined as a crest in which the terminal portion, through strong angulation, comes to be orientated along a different axis from the proximal portion. Thus, a deflecting wrinkle may be described as occurring in relation to any of the principal cusps. On this definition a thick, or well- developed, straight cuspal ridge is not recognized as a deflecting wrinkle. In the present study, the metaconid was examined for evidence of a deflecting wrinkle. The main crest of this cusp was classified as either (i) absent, (ii) straight and weakly developed, (iii) straight and strongly developed, (iv) deflected, or (v) represented by a distinct accessory cuspulid. FOVEA ANTERIOR AND POSTERIOR The fovea anterior is probably homologous with the trigonid basin (Hrdlicka 1924; Weidenreich 1937; Hershkovitz 1971). The name fovea anterior appears to have been proposed by Selenka (1898). Hrdli¢ka (1924) referred to this feature as the precuspidal fossa and Van Valen (1966) proposed to call it the pre-fossid. Elsewhere, the author (Grine 198la—d) has used the names fovea anterior and trigonid basin interchangeably. In the present study the presence or absence of this fossid was recorded. No distinction of the various configurations assumed by the fovea anterior (Sakura 1979) was made. The fovea posterior, as named by Selenka (1898) occupies the distal or distolingual corner of the occlusal surface. Hrdlicka (1924) referred to this feature as the post-cuspidal basin, but elsewhere in the same paper he used this term in reference to the talonid basin. Van Valen (1966) called this basin the post-fossid, while Hershkovitz (1971) preferred the term post-talonid basin or fossid. Biggerstaff (1968, 1975) utilized the presence or absence of the fovea posterior in combination with the principal cusp number for classifying the lower molar. He recognized three categories of crowns: (i) four-cusped, (ii) five- cusped without a fovea posterior, and (iii) five-cusped with a fovea posterior. It is not clear why Biggerstaff did not recognize a subdivision of four-cusped teeth, namely with or without a fovea posterior. In the course of the present study a number of four-cusped molars with or without a distal fossid were encountered. Hence, Biggerstaff’s (1968, 1975) classification cannot be accepted as complete. The absence or presence of the fovea posterior was recorded here, and no attempt was made to classify the shape or size of the fossid. PERMANENT MOLARS OF NEGRO AND SAN 171 Fig. 8. Lower right permanent molars showing the presence (A) and absence (B) of what was regarded as crenulate enamel in the present study. yD ANNALS OF THE SOUTH AFRICAN MUSEUM ENAMEL WRINKLING The smoothness or crenulation of occlusal enamel has been considered by relatively few workers in studies of recent human dentitions (Pedersen 1949; Senyurek 19525; Kiernberger 1955; Jacob 1967), although the phylogenetic significance of crenulate enamel was the subject of debate in the earlier literature (e.g. Selenka 1898; De Terra 1905; Gregory 1922; Abel 1931; Adloff 1937, 1938; Weidenreich 1937; Gregory et al. 1938). The paucity of population studies on crenulate enamel is understandable for two reasons: (1) it does not appear possible to classify objectively different degrees of enamel folding, and (ii) enamel wrinkling can be studied only on teeth with either no wear or only minimal wear. Moreover, the genetic factors influencing the expression of these wrinkles are uncertain (Biggerstaff 1968), especially in view of the findings by Ockerse (1943), Malherbe & Ockerse (1944), and Moller (1965, 1967) that cuspal height and relief are inversely proportional to the amount of fluorine intake in humans. These workers found that in areas with relatively high fluorine content in the drinking-water, the teeth of residents exhibited low, smooth cusps and shallow, broad primary grooves. Pedersen (1949) recorded the presence of wrinkled enamel in the dentition of the east Greenland Eskimo and provided exampies of what he regarded as crenulate enamel (Pedersen 1949, plate 14, figs 77-78). The Ms; which Pedersen illustrated in his figure 77 is not regarded as crenulate by the present author, but the M, pictured in his figure 78 is considered as such. The difference between crenulate and non-crenulate crowns, as classified here, centres around the degree of development of secondary cuspal crests. Each principal cusp may possess one or more cristids which may be variously represented and delineated. The presence of well-defined secondary cuspal cristids and other subsidiary crests (Fig. 8A) was regarded here as an expression of crenulate enamel. In the present study the absence or presence of crenulate enamel (Fig. 8) was noted in those molars that had suffered only minimal wear. No attempt was made to classify the different degrees of expression of these accessory cristids. RESULTS The frequencies of the principal cusp numbers of the permanent lower molars of the South African Negro and Kalahari San are given in Table 2. No significant sexual differences exist in either racial group in the expression of cusp number on any of the mandibular teeth. The expression of cusp numbers on the first and third molars of the San and South African Negro are essentially similar. About 99,0 per cent of San and Negro first molars are five-cusped. Approximately 80,0 per cent of San and Negro third molars have five cusps. There is a striking difference between these two populations in the number of cusps on the second molar. Whereas the Negro M, shows an almost equal expression of either five or four cusps, the San tooth exhibits five cusps in a high proportion of cases. PERMANENT MOLARS OF NEGRO AND SAN 173 TABLE 2 Lower molar cusp number of South African Negro and San. Cusp M, M, M; Population Sex no. N Ws N Za N Vi MeetOw. S.C Male 5 242 = 99.2 136. 45:0 236 . 78.4 4 2 0,8 166 55,0 61 20,3 3 — — — o- 4 3 Female 5 100 98,0 58 54,7 Ti 80,7 4 2 2,0 48 45,3 17 19,3 3 — — — — — — Combined 5 342 ©9988 194 47,6 307 += 78,9 4 4 ie? 214". S24 qa, 201 3 — _ —_ — 4 1,0 San . ( Male 5 62 100,0 42 80,8 30 = 78,9 4 — — 100, 3192 years ALTE 3 Bee. ae = sa ae ee. Female 5 70 98,6 38. 760 34 = 82,9 4 1 1,4 12%. 24:0 7 i/o 3 a uate ae fe ae she Combined 5 32. 992 80 § 78,4 64 ~#81,0 4 1 0,8 22 D6 15 19,0 3 =e age = cae pae2 TABLE 3 Comparison of lower molar cusp numbers of South African Negro and San reported in the literature. Percentage frequencies. Sexes combined. Population Tooth 5 cusps 4 cusps 3 cusps Reference DOBDGD | Sr By 99,2 0,8 Jacobson 1967 98,8 2 — Present study M, 55,1 44,9 — Shaw 1927 33,6 66,4 — Jacobson 1967 47,6 52,4 — Present study M; 66,4 33,0 0,6 Jacobson 1967 78,9 20,1 1,0 Present study San 5 a a i a M, 100,0 — — Drennan 1929 100,0 — —_— Oranje 1934 100,0 — — Kiernberger 1955 100,0 — — Van Reenen 1966 99,2 0,8 — Present study M, 25,0 75,0 — Shaw 1927 92,5 TES — Drennan 1929 100,0 — — Oranje 1934 44,4 55,6 — Kiernberger 1955 82,4 es — Van Reenen 1966 78,4 21,6 — Present study M3; 86,0 14,0 — Drennan 1929 100,0 — — Oranje 1934 95,0 5,0 — Kiernberger 1955 59,1 40,9 — Van Reenen 1966 81,0 19,0 — Present study 174 ANNALS OF THE SOUTH AFRICAN MUSEUM The difference in cuspal number on the M, was found to be significant between the Negro and San males (X? = 22,66; p < 0,005), Negro and San females (X? = 6,50; p < 0,025) and between the sexually pooled Negro and San samples (X? = 31,30; p < 0,005). Shaw (1927) reported frequencies for the cusp number of the San M, which are almost exactly the opposite of the frequencies recorded here (Table 3). Oranje (1934) stated that all mandibular molars of the San are characterized by five cusps. Unfortunately, neither Shaw nor Oranje stated the provenance of the specimens that they regarded as San. Drennan (1929) recorded cusp number in the dentition of a supposed San population, the remains of which were exhumed in November 1926 from graves near Colesberg, Cape Province (Table 3). Slome (1929) studied the same skeletal remains and noted that ‘these skeletons are the remains of what are supposed to have been Cape Bush people, who died during the smallpox epidemic of 1866. . . . The skeletons from these graves seem to belong to a somewhat mixed group of individuals. Many of them are good Bush types, quite like the Kalahari Bushman... a few are Bantu [sic] in type and there may be Hottentot types. It is possible that skeleton 29 had some European blood’ (Slome 1929: 33-34). Nevertheless, Drennan (1929) included every individual in his ‘San’ sample. Kiernberger (1955), too, recorded the cusp number for a group of Khoisanoid skulls that had been collected in the Kalahari by Poch in 1907-1909. Kiernberger (1955) classified the lower molars according to Janzer’s (1927) method, and thus she did not record cusp number directly. The cusp numbers of the lower molars in her sample can be determined from her data, however, and these are presented in Table 3. The percentage frequencies presented in Table 3 are not the same as those which Kiernberger (1955: 30) reported in her text, as the figures quoted by her are derived from her total sample, which included a range of people variously recorded as ‘Korana, Hottentotten-Buschmannmischlung, Neger-Buschmann- mischlung’. Her data also often includes both teeth of any one type from a single individual. The percentage frequencies of lower molar cusp number recorded here for the Negro and San (Table 2) are comparable with those reported in the more recent surveys of Negro and San dentitions (Table 3), except that there is a disparity between the frequencies reported by Van Reenen (1966) and the present author for cusp number on the San M,: the same material was examined by both authors. It is possible that this discrepancy is the result of interobserver differences in counting cusps, but it is also possible that the disparity is the result of the author’s recomputation of Van Reenen’s (1966) data. Lower molar cusp numbers in the South African Negro, the Kalahari San and other human populations are presented in Tables 4-6. The human first mandibular molar tends to be five-cusped, and the impression gained from Table 4 is that, despite the disparities sometimes evident in the frequencies recorded for single populations, the incidence of five-cusped first molars in the South African Negro and San is relatively high. The second mandibular tends PERMANENT MOLARS OF NEGRO AND SAN E75 TABLE 4 Percentage frequencies of cusp number of lower first permanent molar reported for various human populations. Sexes combined. Group Population 5 cusps 4 cusps Reference Mongoloid Amerindian . . . 100,0 Brewer-Carias et al. 1976 95,6 44 Perzigian 1976 73,4 26,6 Campusano ef al. 1972 100,0 — Dahlberg 1949 89,0 11,0 Devoto & Cigliana 1970 90,0 10,0 Devoto & Perrotto 1972 100,0 — Goaz & Miller 1966 99,3 0.6 Goldstein 1948 80,0 20,0 Hooton 1930 91,3 8,7 Lavelle 1971 100,0 — Leigh 1937 99,4 0,6 Nelson 1938 98,5 i FE Snyder ef al. 1969 Eskimoid . 95,6 45 Goldstein 1931 100,0 — Moorrees 1957 97,7 12 Pedersen 1949 Chinese 100,0 — Hellman 1928 93,2 6,8 Lavelle 1971 Japanese . 91,5 8,5 Matsuda 1961 98,0 2,0 Suzuki & Sakai 1957 99,9 0,0 Turner 1979 Ainu 95,6 44 Suzuki & Sakai 1957 96,8 a Ss Turner 1979 Hawaiian . 98,0 2,0 Chappel 1927 99.0 1,0 Katich & Turner 1975 Easter Island . E 96,0 4.0 Turner & Scott 1976 Wanese -. - 100,0 — Harris et al. 1975 Australoid Austral. Aborigine 97,0 3,0 Campbell 1925 100,0 — Hellman 1928 100,0 — Lavelle 1971 Melanesian Naisoi 80,8 19,2 Bailet et a/. 1968 Naisoi & Kw Oio . 79,0 21,0 Sofaer et al. 1972 New Britain 96,3 3,7 Dahlberg 1961 100,0 0,0 Turner & Swindler 1978 Caucasoid European . 82,5 BES Berry 1976 78,2 2AS Brabant & Twiesselman 1964 84,5 152 Brabant 1971 89.0 11,0 Hellman 1928 83,0 17,0 Hellman 1928 89.4 10.6 Jorgensen 1955 95,0 5,0 Lavelle 1971 American . 86,0 10,0 Dahlberg 1949 89.0 11,0 Hellman 1928 87,0 13,0 Hellman 1928 Yemeni 85.0 15,0 Rosenzweig & Zilberman 1967 Cochini 78,0 22,0 Rosenzweig & Zilberman 1967 Bedouin 83.6 16,4 Rosenzweig & Zilberman 1969 Samaritans 94.3 a | Rosenzweig ef a/. 1969 Tristan da Cunha 88.4 11,6 Thomsen 1955 Negroid Ugandan . 99,7 0,2 Barnes 1969 West African . 99,0 1,0 Hellman 1928 91,7 8,3 Lavelle 1971 East African . 95,1 49 Chagula 1960 South African 98.8 1,2 Present study West Indian . 82.0 18,0 Lavelle 1971 American Negro . 99.0 1,0 Hellman 1928 Khoisanoid San 99.2 0.8 Present study Note that the rows do not in all cases equal 100 per cent. 176 ANNALS OF THE SOUTH AFRICAN MUSEUM TABLE 5 Percentage frequencies of cusp number of lower second permanent molar reported for various human populations. Sexes combined. Group Mongoloid Australoid Melanesian Caucasoid Negroid Khoisanoid Population Amerindian Eskimoid . Chinese Japanese Ainu Hawaiian . Easter Island . Yapese Polynesia . Austral. Aborigine Naisoi New Britain . Melanesian European . American . Yemeni Cochini Bedouin Samaritans : Tristan da Cunha Egyptian . Indian Ugandan. . West African . East African . South African West Indian . American Deere San 84,9 39,1 92,4 71,0 62,0 66,6 72,0 Dimi 32,0 45,9 10,1 63,0 32,6 50,5 52,0 34,9 45,5 24,3 76,8 46,0 77,0 76,6 43,0 55,5 60,9 19,0 20,4 28,8 68,6 42,4 21,0 Ne w 2) » — vise r I -B 00 90 ON 00 SO OMNNUOOC CO Vv ewe we & | | i} ~) v NBENNN SE v Ww SPN MOAN DAR AN ROBDHDOORKAGWAOO ~ oo S cusps 4 cusps Oo OM — BN 00 SO NI N10 B NON 00 W 2.5 SO NI ARON OOPFOWOW OAo v N ve a ~ NOS Naar Oi (Nee Nee ND ANBBAWONDAAIN W v 52,4 75,6 70,0 21,6 3 cusps 3,4 8,3 0,0 0,0 Note that the rows do not in alle cases equal 100 per cent. Reference Brewer-Carias et al. 1976 Campusano et al. 1972 Corruccini 1972 Dahlberg 1949 Devoto & Cigliano 1970 Devoto & Perrotto 1972 Goaz & Miller 1966 Goldstein 1948 Hellman 1928 Hooton 1930 Lavelle 1971 Leigh 1937 Nelson 1938 Perzigian 1976 Snyder et al. 1969 Sofaer, Niswander et al. 1972 Sofaer, Niswander ef al. 1972 Sofaer, Niswander et al. 1972 Sullivan 1920 Hrdligka 1909 Hrdli¢ka 1909 Goldstein 1931 Hellman 1928 Moorrees 1957 Pedersen 1949 Hellman 1928 Lavelle 1971 Matsuda 1961 Turner 1979 Turner & Hanihara 1977 Chappel 1927 Katich & Turner 1975 Turner & Scott 1976 Harris et al. 1975 Suzuki & Sakai 1973 Campbell 1925 Hellman 1928 Lavelle 1971 Bailit et al. 1968 Turner & Swindler 1978 Dahlberg 1961 Sofaer, Maclean et al. 1972 Berry 1976 Brabant 1971 Brabant & Twiesselman 1964 Jorgensen 1955 Lavelle 1971 Hellman 1928 Rosenzweig & Zilberman 1967 Rosenzweig & Zilberman 1967 Rosenzweig & Zilberman 1969 Rosenzweig et al. 1969 Thomsen 1955 Leigh 1934 Sullivan 1920 Barnes 1969 Hellman 1928 Lavelle 1971 Chagula 1960 Present study Lavelle 1971 Hellman 1928 Present study PERMANENT MOLARS OF NEGRO AND SAN 177 TABLE 6 Percentage frequencies of cusp number of lower third permanent molar reported for various human populations. Sexes combined. Group Population 5 cusps 4cusps 3 cusps Reference Mongoloid Amerindian . . 80,1 19,5 Corruccini 1972 80,0 ? v Devoto & Cigliano 1970 83,3 16,6 Devoto & Perrotto 1972 62,1 37,9 Goaz & Miller 1966 51,0 49,0 Leigh 1937 59,4 40,6 Nelson 1938 72,6 27,4 Perzigian 1976 38,5 61,5 Hooton 1930 75,0 24,0 Hellman 1928 99,3 0,6 Goldstein 1948 58,0 42,0 Lavelle 1971 Eskimoid. . . 75,0 25,0 Moorrees 1957 88,9 11,1 Pedersen 1949 52,0 48,0 Hellman 1928 89,9 10,2 Goldstein 1948 Chinese... 42,0 58,0 Lavelle 1971 50,0 50,0 Hellman 1928 Japanese = = = | 87,5 25) Turner 1979 Aimtese 225) 3-4. OAT 35,3 Turner & Hanihara 1977 Hawaiian. . . 86,0 14,0 Chappel 1927 67,0 33,0 Katich & Turner 1975 Easter Island. . 80,0 20,0 Turner & Scott 1976 Yapese 100,0 0,0 Harris et al. 1975 Australoid Austral. Aborigine 73,0 27,0 Campbell 1925 86,0 14,0 Hellman 1928 63,6 36,4 Lavelle 1971 Melanesian Naisoi Hates x 9 S80 42,0 Bailit et al. 1968 New Britain . . 80,0 20,0 Dahlberg 1961 84,4 15,6 Turner & Swindler 1978 Caucasoid European. . . 60,0 39,8 Brabant 1971 52,0 47,0 0,9 Brabant & Twiesselman 1964 60,6 39,4 Jorgensen 1955 35,7 64,3 Lavelle 1971 American . . 38,0 62,0 Hellman 1928 Tristan da Cunha 71,4 28,5 Thomsen 1955 Egyptian . 37,0 57,0 h Leigh 1934 Negroid Ugandan . 60,0 40,0 Barnes 1969 West African . 79,0 20,0 Hellman 1928 61,1 38,9 Lavelle 1971 East African . 58,9 40,7 0,4 Chagula 1960 South African 78,9 20,1 1,0 Present study West Indian . 63,6 36,4 Lavelle 1971 American Negro. 67,0 33,0 Hellman 1928 Khoisanoid San .. . 81,0 19,0 0,0 Present study Note that the rows do not in all cases equal 100 per cent. to be four-cusped (Table 5), and here, too, the incidences of five-cusped second molars in the South African Negro and San are relatively high. The third molar tends to be five cusped in recent human populations, and once again the incidences of five-cusped third molars in the South African Negro and Kalahari 178 ANNALS OF THE SOUTH AFRICAN MUSEUM San appear to be relatively high (Table 6) by comparison with other human populations. The frequencies of the various occlusal patterns on the permanent lower molars of the South African Negro and Kalahari San are presented in Table 7. Chi-square analyses revealed that no significant sexual differences exist in either population group in the expression of the primary occlusal pattern on any of the lower molar types. In the Negro the female tends to have a Y pattern (i.e. a metaconid—hypoconid contact) more often than the male on all molars. The San female tends to evince a Y pattern on the M, somewhat more often than the male, while the frequencies of this pattern on the second and third molars are slightly higher in the San male. . The expression of cuspal contact patterns on the first and third molars of the San and Negro are essentially similar. The San tends to possess a Y pattern more often on the M, and the Negro shows a slightly higher frequency of the + pattern on the M, and X pattern on the Ms. The interracial differences in occlusal pattern configuration on the first and third molars are not statistically significant. However, the difference in primary occlusal pattern expression on the M, was found to be statistically significant between Negro and San males (X? = 22,82; p < 0,005), Negro and San females (X? = 6,68; p < 0,01 with X and Y pattern cells only), and between the sexually pooled San and TABLE 7 Lower molar primary occlusal patterns of South African Negro and San. Cusp M, M, M; Population Sex pattern N A N Y, N Yes Negro Male ¥ 195 79,9 GS 2225 AQ Asks ae 44 =18,0 205— = {6709 L120 3S xX 5 2,0 29 9,6 138 46,0 H — — — — 10 3,3 Female . Y. oy 38).7/ 29 eee: 12” 14.0 + i) 13,3 70 66,6 37 A320 xX 3 Soll Ul 6,6 34) 3S H —- — — — 3 3 Combined . XC PT ASA) OT 2338 52. is = S7/ 16,7 DiS ONE: 149 38,6 xX 8 2,3 36 8,8 172 44,6 H — — — — 13 3,4 San Male ye 57 83,9 Di S12) 9 231 + I@ iG. 24 46,2 14-3658 xX — — 1 1,9 14. 3638 H — — — = 1 2,6 Female . Y 66) > 93.0 25 50,0 7 17. = 5 7,0 24 48,0 17» sags xX — — 1 2,0 15 36% H — = = = 2 4,9 Combined . Y ISS S35 7/ See SiO) 16. > 202 a 15 11,3 48 47,0 31 392 xX — 2 2,0 29-3637 H — — = = 3 3,8 PERMANENT MOLARS OF NEGRO AND SAN 179 Negro samples (X? = 31,16; p < 0,005). The San possesses a Y pattern on the second molar in a significantly higher percentage of cases than the Negro, while the Negro molar more often shows a cruciform (+) type of cuspal contact (Table 7). There is also a striking difference between these two racial groups in the number of cusps on this molar. In both the San and the Negro the first molars show the highest frequencies of the Y pattern with the second and third molars showing progressively lower frequencies of this pattern. The second molars of both populations show an increased frequency in the + pattern over the first molars, while the third molars evince a X pattern more often than the second (or first) molars. The H pattern was encountered only on the third molars of both popula- tions. The frequency of expression of this variant pattern is low and almost identical in the San (3,8 per cent) and the Negro (3,4 per cent). Van Reenen (1966) and Jacobson (1967) have reported data on the fre- quencies of the primary occlusal pattern in San and South African Negro populations respectively. Although neither stated the scheme by which they classified the occlusal pattern, it is apparent from Jacobson’s illustrations and, since they recognized only the Y and + patterns, that Hellman’s (1928) standard was employed by both. The frequencies reported by them are compared with the incidences of Y and + (where + consists of combined + and X pattern frequencies) recorded in the present study in Table 8. Both Van Reenen (1966) and Jacobson (1967) recorded higher frequencies of the Y pattern on the M,, and lower frequencies of this pattern on the M, and M, for the San and Negro than the frequencies reported here. In no instance, however, are the incidences reported by Van Reenen (1966), Jacobson (1967), and the present author markedly different. * TABLE 8 Percentage incidence of primary occlusal patterns of lower permanent molars recorded for Negroid and Khoisanoid populations. Sexes combined. M, M, M; Population ts yA = iV IF Reference Bieondan= 2 |. . = 83,8 16,2 32.5) ORS 19,2 80,8 Barnes 1969 Bast African. . ...: 86,9 13,1 ALVIS WTB 2ST IIT Chagula 1960 West African . . . 100,00 — 2910, 70 23,0 76,0 Hellman 1928 Sos hS9 250 SO 30,6 69,4 Lavelle 1971 SouemAtmican . ... 89,8 10,2 123 87 Ut LIAS Jacobson 1967 Pete i 1 SiO) 190 23,8 16,2 15558655 Present study Pechicais, < <2. «+ ~ 99,0 1.0 DLO SO Oe 93:0 Hellman 1928 Miestindian . << . . 83,6 16,4 30,0 70,0 32,4 67,6 Lavelle 1971 Kalahnatr san... . 981 1,9 525586755 13,6 86,4 Van Reenen 1966 ** S37) ts 51,0 49,0 202 1938 Present study *Frequencies of + pattern obtained by combining data for + and X patterns for comparison. Comparison of the data for the South African Negro and San with various Negro populations (Table 8) indicates that the first and third molars of both the San and South African Negro and the second molars of the South African 180 ANNALS OF THE SOUTH AFRICAN MUSEUM Negro are similar to other Negro populations in the relative frequency of occlusal pattern expression. The M, of the San, according to the present data, tends to have a Y pattern more often than do any of the Negro populations. According to Van Reenen’s (1966) results, the frequency of the Y pattern on the M, of the San is higher than in all other Negro populations except the Teso of Uganda (Barnes 1969). Kiernberger (1955) examined the occlusal surface pattern of the Khoi- sanoid lower molar, but she utilized Janzer’s (1927) method of classification. Janzer (1927) recognized three different occlusal types, the foundation of each being cusp number and occlusal outline. Although Kiernberger (1955: 25) states that “in die dritte Gruppe der unteren . . . Molaren geh6ren diejenigen, TABLE 9 Percentage frequencies of primary occlusal patterns of lower first permanent molar reported for various human populations. Sexes combined. These authors have followed Hellman’s (1928) method. Group Population Y = Reference Mongoloid AMenringiani ea eee 73,4 26,6 Campusano et al. 1972 100,0 — Hellman 1928 63,0 37,0 Corruccini 1972 99,4 0,6 Dahlberg 1949 69,3 30,6 Goldstein 1948 88,4 11,6 Lavelle 1971 88,6 11,4 Nelson 1938 84,0 16,0 Snyder et al. 1969 Jase! 4 4 5 < 91,1 9,0 Goldstein 1948 97,0 3,0 Hellman 1928 41,4 58,6 Moorrees 1957 94,7 4,1 Pedersen 1949 GhineSée is) ont oe ee 92,4 7,6 Lavelle 1971 100,0 — Hellman #928 Australoid Austral. Aborigine . . 100,0 — Hellman 1928 Ofel 2,3 Lavelle 1971 Melanesian Melanesian. . . . 96,8 3,2 Dahlberg 1961 60,1 39,8 Lombardi 1975 Caucasoid EUhOpedit. nen 46,0 Spi Brabant 1971 85,0 15,0 Brabant & Twiesselmann 1964 79,0 21,0 Lavelle 1971 PONTING 5 8 92,0 4,0 Dahlberg 1949 94,0 6,0 Hellman 1928 Yenre ya Seen a 53,0 47,0 Rosenzweig & Zilberman 1967 Cochin.) Vie eS: 57,0 43,0 Rosenzweig & Zilberman 1967 Bedouink. ~. @.o . = 70,4 29,6 Rosenzweig & Zilberman 1969 Samaritans). 4) 472 87,1 12,9 Rosenzweig et al. 1969 Tristan da Cunha . . 74,7 PS) Si Thomsen 1955 Negroid EasteAtnical.. aaa 86,9 13,1 Chagula 1960 West Africa. . . . 100,0 — Hellman 1928 86,1 13,9 Lavelle 1971 SouthyAtiricas = en ae 89,8 10,2 Jacobson 1967 American Negro. . 99,0 1,0 Hellman 1928 West Indian. . . . 83,6 16,4 Lavelle 1971 Khoisanoid= “San? -i2= 2 0) 3 a 98,1 1,9 Van Reenen 1966 Note that the rows do not in all cases equal 100 per cent. PERMANENT MOLARS OF NEGRO AND SAN 18] TABLE 10 Percentage frequencies of primary occlusal patterns of lower first permanent molar reported for various human populations. Sexes combined. These authors have followed Jorgensen’s (1955) method. Group Population ¥ == 4 Reference Mongoloid Amermdian <: . . d,6 29,4 19,0 Brewer-Carias et al. 1976 65,0 35,0 0,0 Devoto & Cigliano 1970 69,2 DBM EL Goaz & Miller 1966 Z 14,2 2 Sofaer et al. 1972 Chinese See eee is ? ? 4,1 Turner 1976 Japaneser= 2° 7 = y7iles 22,9 5,8 Matsuda 1961 ? k 6,6 Turner 1976 70,7 227 Gof Suzuki & Sakai 1957 L 1k Se Turner 1976 88,9 7,9 32 Turner 1979 ASHES he ea 2) 593.4 33 333 Turner & Hanihara 1977 56,5 34,8 8,7 Suzuki & Sakai 1957 Relvnesian .. 4 =] 2 82.4 1283 3,3 Suzuki & Sakai 1973 85,0 ? 2 Turner & Scott 1976 Hawaiian . . . . -890 ? Katich & Turner 1975 Wanese? aft 2 8s 730 ? L Harris et al. 1975 Melanesian Melanesian) .*.. ? 33,0 ? Sofaer et al. 1972 INGISOlas) ses 2 (SAD 45,1 0,0 Bailit et al. 1968 INewsBritain. <9 .2-.. 9207 73 0,0 Turner & Swindler 1978 Caucasoid EBEODEAN uae =. 4» 9959 34,6 9,9 Berry 1976 69,2 1357 EZ! Jorgensen 1955 69,2 18,1 27 Jorgensen 1955 American... = «+. 298 59,7 10,5 Takehisa 1957 Negroid Weandan, .. .. =. .: ~ $3.8 16,2 o Barnes 1969 South Africa: : «. « 82,0 16,7 2S Present study Maeisanoid Sait: . .:. . » .~. 88,7 L3 0,0 Present study die vier Hocker besitzen und deren Furchen Kreutzf6rming angeordnet sind’, the alignment of the mesiobuccal and lingual grooves, rather than cuspal contact, appears to be the basis for the cruciform (Kreutzf6rming) pattern. Thus, Kiernberger’s data on the occlusal morphology of the Khoisanoid lower molar do not seem to be comparable with the primary occlusal pattern data recorded here. The majority of authors who have reported frequencies of primary occlusal patterns for different human populations have utilized either Hellman’s (1928) or Jorgensen’s (1955) methods of classification. As pointed out above, these two systems are not directly comparable in all instances as Hellman’s + type includes both the + and X patterns of Jorgensen. The data presented by various authors for different populations are tabulated in Tables 9-14, according to the classificatory method employed by each author. Several workers have created and/or utilized methods of classification that are either slight modifications of, or completely different from, either the Hellman or Jorgensen systems. It is essential that these various classifications be recognized as distinct when one attempts to compare data on different population groups recorded by various authors. Numerous workers have, nevertheless, compiled rather extensive 182 ANNALS OF THE SOUTH AFRICAN MUSEUM tables of reported occlusal pattern frequencies in which the incomparability of much of the reported data has been ignored (e.g. one list included studies that had used variously Hellman’s (1928), Jorgensen’s (1955), and Steslicka’s (1948) categories). Tables 9-14 indicate that the human first permanent molar tends usually to have a Y pattern, while the second and third molars evince the + or X pattern more frequently than the Y. In those studies that considered the X pattern, the Ms possesses the highest frequency, the M, a somewhat lower, and the M, the lowest frequency of this pattern (Tables 10, 12, 14). It is evident also that, in general, the expression of the primary occlusal pattern is more variable than the principal cusp number for each molar type. Examination of Tables 9-14 indicates that, while there is a considerable degree of intragroup variability in the frequencies of expression reported TABLE 11 Percentage frequencies of primary occlusal patterns of lower second permanent molar reported for various human populations. Sexes combined. These authors have followed Hellman’s (1928) method. Group Population Y ae Reference Mongoloid Amerindian ene 39,1 60,9 Campusano et al. 1972 4,5 95,5 Corruccini 1972 3,0 97,0 Dahlberg 1949 4,9 95,1 Goldstein 1948 1,0 99,0 Hellman 1928 5,8 94,2 Lavelle 1971 9,6 90,4 Nelson 1938 38,5 61,5 Snyder et al. 1969 Eskimoid Si, eae 15,8 84,3 Goldstein 1948 13,0 87,0 Hellman 1928 0,0 100,0 Moorrees 1957 24,3 76,6 Pedersen 1949 Chinese 2 |: hae: 0,0 100,0 Hellman 1928 15,6 84,4 Lavelle 1971 Australoid Austral. Aborigine. . 5,0 95,0 Hellman 1928 pel 77,3 Lavelle 1971 Melanesian Melanesian. .. . 12,8 87,2 Dahlberg 1961 Caucasoid Europedie ee 33,1 66,6 Brabant 1971 9,5 90,5 Brabant & Twiesselman 1964 Teal 92,9 Lavelle 1971 American a 84,6. 32 5,0 95,0 Hellman 1928 YEMEN 5 cise 9,0 91,0 Rosenzweig & Zilberman 1967 Gochinil< -% . 4 47: 5,0 95,0 Rosenzweig & Zilberman 1969 Bedouin. 920-2 2% 7,0 93,0 Rosenzweig & Zilberman 1969 Samanritam 7).° 29°22. 12,6 87,4 Rosenzweig et al. 1969 Tristan da Cunha . . 7,9 92,1 Thomsen 1955 Negroid East African . . . 20,2 79,8 Chagula 1960 West African’ = 5°". 29,0 71,0 Hellman 1928 1» 20 75,0 Lavelle 1971 South African . . . 12,3 87,7 Jacobson 1967 AIMebICae 2) 7 eae 6 27,0 73,0 Hellman 1928 West Indians.) 5 25>. 713010 70,0 Lavelle 1971 Khorsanoid>” VSan 7 sn oe 32,5 67,5 Van Reenen 1966 PERMANENT MOLARS OF NEGRO AND SAN 183 TABLE 12 Percentage frequencies of primary occlusal patterns of lower second permanent molar reported for various human populations. Sexes combined. These authors have followed Jorgensen’s (1955) method. Group Population iG = X Reference Mongoloid Amerindian o . « - 5,6 34,8 59,6 Brewer-Carias et al. 1976 — 45,0 55,0 Devoto & Cigliano 1970 OW 53,3 36,0 Goaz & Miller 1966 Japanesen we aa.) 4,2 41,0 54,7 Matsuda 1961 aS 44,2 S13 Suzuki & Sakai 1956a SHA5) 28,8 39,7 Turner 1979 NII eee 2 ea FATS 32,1 56,6 Turner & Hanihara 1977 Polynesian. . . . 11,0 ? 2 Turner & Scott 1976 10,1 38,0 51,9 Suzuki & Sakai 1973 Hawaiian 2) =... 2050 v ? Katich & Turner 1975 Nielanesiam’ " Naisoi . . . . . ~*~ 13,1 86,9 — Bailit et al. 1968 New Britain... 8,1 67,6 24,3 Turner & Swindler 1978 Melanesian . . . ? 80,0 2 Sofaer, Maclean ef al. 1972 Caucasoid European 2) .) -.) 1429 54,6 30,5 Berry 1976 16,7 26,8 56,6 Jorgensen 1955 14,3 Died 58,5 Jorgensen 1955 American... Lo. 42,1 57,9 Takehisa 1957 Negroid Weandanvesn 8. 12. 32.5 60,0 WS Barnes 1969 SouthyAirican,. . =. 623.8 67,4 8,8 Present study Kinoisanoids “Sam... . .. . ..: 31,0 47,0 2,0 Present study (e.g. frequencies for the Y pattern on the M, of Mongoloid peoples range between 41,4-100,0 per cent), overall the Australoid peoples tend to possess the highest frequencies of Y-patterned first molars followed closely by the San and the Negro. The Caucasoid populations that have been examined tend to show the lowest frequencies of the Y pattern on the M,. Conversely, the Caucasoid populations tend to display the highest frequencies of + patterns and/or X patterns on the first mandibular molars. The percentage frequencies of Y patterns on second molars that have been recorded for various human populations range from total absence to 51,0 per cent (Tables 11-12). Van Reenen (1966) recorded that 32,5 per cent of San M,’s possess a Y pattern, while the present author found 51,0 per cent of these molars to be Y patterned. The latter figure represents the highest incidence recorded for any human population, and Van Reenen’s figure is amongst the highest of recorded frequencies (Tables 11-12). It appears, therefore, that the Y pattern is Shown more frequently on the M, of the San than on this tooth in most other human populations. Whereas the Melanesian and Australoid peoples tend to possess the highest frequencies of Y-patterned first molars, the second molars of the populations appear to show relatively low frequencies of Y pattern expression. There seems to be little difference in the incidence of Y pattern expression on the M, amongst the Mongoloid, Australoid, Caucasoid, and Negroid populations (Tables 11-12). 184 ANNALS OF THE SOUTH AFRICAN MUSEUM Whereas the San tends to have the Y pattern relatively frequently on the M, and M,, the incidences of this pattern on the Mg as reported by both Van Reenen (1966) and the present author are amongst the lower recorded for other human populations (Tables 13-14). The frequencies of Y-patterned M,’s in the South African Negro recorded by both Jacobson (1967) and the author are the lowest recorded for any other Negroid population with the excep- tion of the American Negro (Tables 13-14). The frequencies of Y-patterned third molars in the Melanesian and Australoid populations are amongst the lower percentage incidences recorded for recent human populations. However, comparison of only that data recorded by authors who used Jorgensen’s (1955) method of crown classification (Table 14) indicates that the San Mg, shows a Y pattern more frequently than in any other population (the Teso of Uganda and the Ainu show comparable frequencies though), and that the frequency of X-patterned M,’s in this population is one of the lowest recorded to date. Chi-square evaluation of cuspal number and primary occlusal patterns on the lower molars of the South African Negro and San revealed that the two are dependent variables only on the M, of the Negro male (X? = 6,47; p = 0,04) TABLE 13 Percentage frequencies of primary occlusal patterns of lower third permanent molar reported for various human populations. Sexes Combined. These authors have followed Hellman’s (1928) method. Group Population Y =F Reference Mongoloid Amerindian. .. . Ul! 92,3 Corruccini 1972 29,0 71,0 Dahlberg 1949 18,7 81,4 Goldstein 1948 5,0 94,0 Hellman 1928 18,8 78,2 Lavelle 1971 13,2 86,8 Nelson 1938 Eskimoideis cee 20,4 79,7 Goldstein 1948 — 100,0 Hellman 1928 — 100,0 Moorrees 1957 23,8 76,2 Pedersen 1949 ChineSse'sc- 24 Sgt OA 38 — 100,0 Hellman 1928 43,4 56,6 Lavelle 1971 Australoid Austral. Aborigine. . 14,0 86,0 Hellman 1928 18,2 81,8 Lavelle 1971 Melanesian Melanesian. . . . 16,6 83,4 Dahlberg 1961 Caucasoid European . «i. . 43,4 56,4 Brabant 1971 30,4 69,5 Brabant & Twiesselman 1964 6,8 93,2 Lavelle 1971 American «) @s 50% 4,0 96,0 Hellman 1928 Tristan da Cunha . . 71,4 28,5 Thomsen 1955 Negroid East African... 20,3 79,7 Chagula 1960 West African... 23,0 76,0 Hellman 1928 30,6 69,4 Lavelle 1971 South African... 7,4 92,6 Jacobson 1967 AINenICaN) 9h) ayo 4 7,0 93,0 Hellman 1928 West Indian te hea) 32,4 67,6 Lavelle 1971 Khoisanoid’) «Sant. Fie ee 13,6 86,4 Van Reenen 1966 PERMANENT MOLARS OF NEGRO AND SAN 185 TABLE 14 Percentage frequencies of primary occlusal patterns of lower third permanent molar reported for various human populations. Sexes combined. These authors have followed Jorgensen’s (1955) method. Group Population Y + X References Mongoloid Amerindian . . . 6,9 48,2 44,8 Goaz & Miller 1966 0,0 0,0 100,0 Devoto & Perrotto 1972 Japanesea ae ee | MES 13,7 74,5 Turner 1979 PANITIUG Ses Coe eS La S5i/ Tel Turner & Hanihara 1977 Melanesian IN@ISOMe fo ee aa 81,0 11,9 Bailit et al, 1968 New Britain... 0,0 222 77,8 Turner & Swindler 1978 Caucasoid European a). 2) 2 2 1059 A 76,9 Jorgensen 1955 ie? 17,5 Ws Jorgensen 1955 Negroid Weandan- . . =. . 19,2 46,1 34,3 Barnes 1969 South African. . . 13,5 38,6 44,6 Present study Khotsanoid Sane eeesies a: 3h 320.2 39,2 S17) Present study and on the M, of the Negro female (X? = 15,24; p = 0,005). No evidence of variable dependence was found for any of the other Negro molars, nor were cusp number and occlusal pattern found to be dependent in any case in the San. Studies of these two features on the lower molars of other human popu- lations have also found them to be unrelated (Jorgensen 1955; Garn et al. 1966a, 19666; Devoto & Cigliano 1970; Devoto et al. 1970) or only very slightly related within individuals (Rosenzweig & Zilberman 1969). Nevertheless, numerous workers have recorded the cusp number and occlusal pattern together (e.g. Y5, Y4, +5, +4, etc.). In order to facilitate comparisons with the results presented in such studies, the frequencies of cusp number-occlusal pattern ‘complements’ on the South African Negro and San lower molars are presented in Table 15. The first molars of both the Negro male and female most frequently evince a Y5 pattern. This is true also of the San first molar. The third molars of both the Negro and San males show an XS pattern most frequently, while the third molars of the Negro and San females exhibit most commonly a +5 pattern. However, in the sexually pooled samples the Negro M, is found to display most frequently an X5 pattern (because of the larger sample of male molars) but the San M, is found to evince a +5 pattern most often. Both the Negro male and female show a +4 pattern most frequently on the second molar (Table 15). The San male has a Y5 pattern most commonly, but this frequency is only slightly higher than that for the +5 pattern. The YS and +5 patterns occur with the same frequency on the second molars of the San female, and both these patterns are more common than any other. In the sexually pooled San sample the Y5 pattern is found most frequently on the second molar, with the +5 pattern encountered in only a slightly lower per- centage of cases. The incidences of metaconid deflecting wrinkles (as this feature is here defined) on the lower molars of the South African Negro and San are presented in Table 16. ANNALS OF THE SOUTH AFRICAN MUSEUM 186 worn FTNoTFTNN TH TH THN TH THTH Seth Kh Oe ett ch K Oe td KK ‘Ou udsajjod dsny dsny uDS' 1ADYDIDY OABIN] UDALL YINos ‘ues LIeYR[ey pue OIBONY ULOTIPY YINOG oy} JO UsJO}Jed [edsno pue Joquinu dsno IejOUT JIOMO'T C] alavy pourquioy o[ewo.y ' Ie xas’ PERMANENT MOLARS OF NEGRO AND SAN 187 TABLE 16 Metaconid crest morphology of lower permanent molars of the South African Negro and Kalahari San. Population Absent Weak Stron Deflected Cuspuli and sex Tooth IN = IN 9h, No INCE le N Ss Negro RABIES M, ES 6:3 13307 1249-1 38 15,9 0 0,0 M, 120) 40-2 isi sle9 14 48 Oe 7A 0) 0:0 M; 122 41,8 143 49,0 14 4,8 | i O38 Remale =. . M, De DAY) Slee 56" 5616 26) 2633 0 0,0 M,; 41 39,4 3529 ee eT 0 0,0 VY OG) M; 37 42,0 43 48,9 3) 3.4 Dee! 0 0,0 Combined. . M, ii 0 88 26,1 168 49,9 64 19,0 0 0,0 M, 161 40,8 206 52,2 apy NAS, Gy) ales YW OW M; 159 41,8 186 49,0 L745 17 4,4 103 San Migiepa. 4 . M, 0 0,0 13, 2120 36 58,1 13 20,9 0 0,0 M, 2-39 39 76,5 8 15,7 PH By) 0-050 M; 7 18,4 23 60,5 7 18,4 2G 0 0,0 bemaie . . M, O7010 7 10 46 67,6 Sa 0 0,0 M, 4 8,2 BY2" (Sp ES: 10 20,4 a ull 0 010 M; 8 21,6 24 64,9 Dar Bye 3) teal OF O:0 Combined. . M, 0” O00 20 15,4 82 63,1 Pes PANES) 0 0,0 M; 6 6,0 Te Wy 18 18,0 55,0 0 0,0 M; 1577 20:0 47 62,7 Bey) Au 5.3 0 10:0 No significant sexual difference in the expression of metaconid cristid morphology is shown by the San, and in the Negro no statistically significant sexual difference in metaconid cristid morphology is shown on the second and third molars. On the first permanent molar, however, the Negro female shows a significantly higher frequency of deflecting wrinkles than does the male Ve 452: p < 0,05). In both the Negro and the San the principal cristid of the metaconid of the first molar is most commonly well developed and straight. The main meta- conid cristids of the second and third molars of both populations are most frequently weakly developed and are not deflected (Table 16). The frequency of deflected metaconid cristids in both sexes of both popu- lations is considerably higher on the first molar than on either the second or third molars. In the Negro (both sexes) and in the San female the metaconid crest of the second molar is deflected somewhat less frequently than on the third molar. In the San male the frequency of deflecting wrinkles on the M, is slightly higher than on the Ms. In the sexually pooled San sample the second and third molars show nearly identical frequencies of deflected metaconid cristids. There is no significant difference between the San male and the Negro male in the frequencies of deflecting wrinkles on any of the lower molars, nor is there any significant difference between the San female and Negro female frequencies on the first and third molars. However, whereas no Negro female M, was found to possess a deflecting wrinkle, this feature was expressed on 188 ANNALS OF THE SOUTH AFRICAN MUSEUM some 6,1 per cent of San female second molars (X? = 6,49; p < 0,025). Com- parisons of sexually pooled data indicates that no significant difference exists between the South African Negro and the Kalahari San in the frequency of expression of deflecting wrinkles on either the first or third molars. The San possesses a deflecting wrinkle on the second molar in a significantly higher percentage of cases on the second mandibular molar (X? = 4,45; p < 0,05). Morris (1970) noted that the presence of a metaconid deflecting wrinkle may effect a metaconid—hypoconid contact. In the Negro male a deflecting wrinkle and a metaconid—hypoconid contact (Y pattern) were found together in only 16,2 per cent of first molars, 1,5 per cent of second molars, and 7,7 per cent of third molars. These two features were found together on the first, second and third molars of the Negro female in 24,4 per cent, 0,0 per cent, and 8,3 per cent of cases respectively. In the San male they were encountered together on 21,2 per cent of first, 7,4 per cent of second, and 0,0 per cent of third molars. The incidences of concomitant expression of these two characters in the San female molars are similar to the frequencies encountered on the San male teeth. Suzuki & Sakai (19655), in a study of the deflecting wrinkle on the first lower molars of recent Japanese, recorded that this character was present in 49,3 per cent of male and 23,8 per cent of female teeth exhibiting a Y pattern. It is clear that there is no interdependence of the deflecting wrinkle and the Y pattern on the lower molars of the South African Negro and the San. How- ever, aS pointed out above, Morris (1970), amongst others (e.g. Hanihara 1961 ; Hanihara et al. 1975; Turner & Hanihara 1977; Turner 1976, 1979; Turner & Swindler 1978), has erroneously identified a strongly developed, straight meta- conid crest as a deflecting wrinkle. A highly significant (p < 0,005) relationship between occlusal pattern and metaconid crest morphology was found on all the lower molars of both sexes of the Negro and San, but the Y pattern (a meta- conid—hypoconid contact) was found to be dependent upon the presence of a large, straight metaconid crest. Reported frequencies of deflecting wrinkles on the lower molars of various human populations are tabulated in Table 17. It is apparent that the majority of authors have classified both straight, well-developed metaconid crests and truly deflected cristids as deflecting wrinkles. Morris (1970) reported that 46,3 per cent of South African Negro and 78,5 per cent of San first molars show a deflecting wrinkle. By combining straight, well-developed crests and deflected metaconid cristids into a single category —‘deflecting wrinkle’-—as advocated by Morris (1970) and others, some 68,9 per cent of Negro and 84,6 per cent of San first molars examined by the present author were found to evince this feature (Table 17). The figures (combined) are somewhat, albeit insignificantly, higher than those reported by Morris (1970) for the same populations. Comparison of the results obtained by authors who have followed this method of classification (Table 17: fre- quencies preceded by x) reveals that only the Amerindian (Morris 1970), PERMANENT MOLARS OF NEGRO AND SAN 189 Chinese (Turner 1976), and Wajin (Hanihara ef al. 1975) show a ‘deflecting wrinkle’ on the first molar more commonly than does the South African Negro. Only the Chinese (Turner 1976) show a higher frequency of this character than the San on the M,. TABLE 17 Incidence of metaconid deflecting wrinkle on lower permanent molars reported for various human populations. Sexes combined. Group Population M, M, M; Reference Mongoloid Amerindian * 74,8 Morris 1970 Japanese 29,6 Hanihara ef al. 1964 23,4 Suzuki & Sakai 19565 z SEG Turner 1976 = 54,2 0,0 0,0 Turner 1976, 1979 Ainu e Sl Turner 1976 = Sil 10,7 3,6 Turner & Hanihara 1977 Chinese 2 87,5 Turner 1976 Wajin = 71,6 Hanihara et al. 1975 Melanesian New Britain * 60,0 5,4 0,0 Turner & Swindler 1978 Caucasoid European * 34,2 Axelsson & Kirveskari 1977 South African* 7,0 Morris 1970 Indian eS Morris 1970 Negroid South African* 46,3 Morris 1970 . 68,9 al 8,9 Present study 19,0 15 4,4 Present study Khoisanoid San = 78,5 Morris 1970 84,6 23,0 17,3 Present study PS 5,0 5,3 Present study * Frequencies include straight, well-developed metaconid crest as well as true deflecting wrinkles. If one considers only those studies in which a truly deflected metaconid crest has been classified as a deflecting wrinkle, the Negro and San frequencies of this character are considerably lower (19,0 and 21,5 per cent respectively), as are the incidences reported for Mongoloid peoples (Suzuki & Sakai 1956b; Hanihara et al. 1964) (Table 17 herein). The frequencies of true deflecting wrinkles on the first lower molars of the South African Negro and San are only slightly lower than those reported for Japanese populations. It has been suggested (Hanihara 1963, 1966, 1967, 1968a, 19685, 1970; Hanihara et a/. 1964) that the deflecting wrinkle be included in the ‘Mongoloid dental complex’ of both the deciduous and permanent teeth. From the foregoing it is apparent that, regardless of whether a straight, strongly developed meta- conid crest and a deflected crest or only a truly deflected metaconid crest is classified as a deflecting wrinkle, the frequencies of this feature on the San and Negro molars are within the range of frequencies recorded for Mongoloid peoples. Axelsson & Kirveskari (1977) have suggested that the inclusion of the deflecting wrinkle in the ‘Mongoloid dental complex’ be re-evaluated. It is evident that the racially diagnostic value of this feature in itself is doubtful, 190 ANNALS OF THE SOUTH AFRICAN MUSEUM but it forms only part of the ‘Mongoloid dental complex’ as defined by Hanihara. More particularly, this complex consists of concomitantly high expressions of (1) shovelled incisors, (ii) tuberculum intermedium on the dm, and M,, (iii) tuberculum sextum on the dmz, (iv) protostylid on the dmg, (v) plagioconule (metaconule) on the dm?, and (vi) deflecting wrinkle on the dm, and M,. The high frequencies of one of these characters (e.g. a deflecting wrinkle) in non- Mongoloid populations does not necessarily detract from its significance as a part of a complex of other features. The frequencies of the tuberculum sextum recorded for the permanent lower molars of the South African Negro and San are presented in Table 18. No significant sexual dimorphism in the possession of this accessory cuspulid is present in either the San or Negro samples examined. TABLE 18 Incidence of tuberculum sextum on lower permanent molars of the South African Negro and Kalahari San. M, M, Ms; Population Sex N V/A N o/s N Vs Negro Maller es. 9). 12 52 sy Syl 58 19,9 Remale 9%. 6 §66,4 y, 1,9 19° 2251 Combined. . es SS) iy 4,3 77 204 San Malate ek. 10 16,4 6 14,3 6 20,7 Female ~.. . 10 15,4 Sl 14 38,9 Combined. . 20 15,9 a se2 20 30,8 While the Negro and San males show a tuberculum sextum on the third molars in a nearly identical percentage of cases, the San male possesses this cuspulid significantly more frequently on the first molar (X? = 8,83; p < 0,005) and the second molar (X? = 5,29; p < 0,025). The San female tends to possess a tuberculum sextum more frequently than the Negro female on all three mandibular molars and, while the occurrences are not significantly different on the first and third molars, the San female has this cuspulid in a significantly higher frequency on the second molar (X? = 6,75; p < 0,01). Comparison of the sexually pooled data reveals that the San shows a significantly higher incidence of the tuberculum sextum on the first molar (X? = 12,72; p < 0,005) and second molar (X? = 10,11; p < 0,005). The San also tends to show this cuspulid more frequently than the Negro on the third molar, but the difference in this instance was found to be statistically not significant. Jacobson (1967) recorded the frequency of a tuberculum sextum on the permanent lower molars of the South African Negro, but he did not mention the criteria by which he classified this cuspulid. His figures for the occurrence of the tuberculum sextum in this population are consistently (albeit insignifi- cantly) lower than those in this study for all the molars. The differences amongst these figures may reflect different sample sizes, different methods of classifi- PERMANENT MOLARS OF NEGRO AND SAN 191 cation, or the fact that Jacobson (1967) scored teeth from both sides of the same jaw in a number of instances. Kiernberger (1955) recorded the presence of this cuspulid in her Khoisanoid sample, but she did not differentiate between it and the tuberculum intermedium. She combined the occurrence of either one or both into a single molar classification. Frequencies of the tuberculum sextum recorded by other workers for the lower molars of various human populations are presented in Table 19. This TABLE 19 Percentage frequencies of tuberculum sextum on permanent lower molars reported for various human populations. Sexes combined. Group Population M, M, Ms; Reference Mongoloid Amerindian . . 49,0 14,5 Brewer-Carias et al. 1976 6,0 16,0 32,0 Hellman 1928 3.3 2,6 4,8 Nelson 1938 Bskimoid) .« . . 1,0 20,0 28,0 Hellman 1928 72 10,3 16,0 Pedersen 1949 Chinese ie ea. 0,0 0,0 0,0 Hellman 1928 31,0 Turner 1976 wapanese = = —. Dare dl Hanihara 1970 18,4 Suzuki & Sakai 1957 5,6 1,6 Takehisa 1957 25,3 Turner 1976 58,0 Turner 1976 58,0 14,0 Die Turner 1979 PMU ewe cory ae. 17,8 Hanihara 1970 PANT) Suzuki & Sakai 1957 29,8 Turner 1976 25,8 8,5 9,1 Turner & Hanihara 1977 Polynesian. . . 54,5 30,0 50,0 Suzuki & Sakai 1973 Hawatian ..: : . 43,0 2,0 4,0 Katich & Turner 1975 Easter Island . . 19,0 9,0 14,0 Turner & Scott 1976 NIADESEs a 8 710 77,0 80,0 Harris ef al. 1975 Australoid Austral. Aborigine 8,0 43,0 54,0 Hellman 1928 Melanesian Melanesian ae 9,0 5,0 9,0 Dahlberg 1961 New Britain... 325 9,8 24,2 Turner & Swindler 1978 Caucasoid European 2 2% 17,0 10,9 Axelsson & Kirveskari 1979 INMEHCAN ©. *-2> <2 0,0 0,0 3,0 Hellman 1928 0,0 0,0 Takehisa 1957 Negroid East African... 6,6 0,3 2,0 Chagula 1960 West African... 3,0 6,0 16,0 Hellman 1928 South African. . 5) 4,3 20,4 Present study AMIMETICANL 5 osc. =: 2,0 8,0 33,0 Hellman 1928 Kboranoid San. . . . .- 15,9" 12 30,8 Present study table does not include all the references for recent human samples as some workers have combined the presence of a tuberculum sextum and a tuberculum intermedium into a single category (e.g. Goldstein 1931; Berry 1976). Others have recorded the presence of cusp numbers over five but have not stated whether the tuberculum sextum was, indeed, always the sixth cusp (e.g. Gold- stein 1948; Thomsen 1955; Matsuda 1961; Brabant & Tweisselmann 1964; 192 ANNALS OF THE SOUTH AFRICAN MUSEUM Devoto & Cigliano 1970; Brabant 1971; Devoto & Perrotto 1972; Perzigian 1976). Examination of Table 19 indicates that the first lower molars of the South African Negro possess a tuberculum sextum about as often as the M,’s of the east African Negro and slightly more frequently than the molars of the west African Negroes. The frequency of the presence of the tuberculum sextum on Negro first molars tends to be relatively low. The San first molars exhibit this cuspulid more frequently than do the Negro teeth. Whereas Hellman (1928) and Takehisa (1957) recorded a complete absence of this cuspulid in their American Caucasoid first molar samples, Axelsson & Kirveskari (1979) reported that 17,0 per cent of Icelandic Caucasoids examined by them possessed a tuber- culum sextum on the first molar. Sometimes dramatic differences in the frequency of the tuberculum sextum on the first lower molars of various Mongo- loid peoples have been recorded (Table 19). However, the incidences of this cuspulid in a number of Mongoloid populations are considerably higher than those recorded for Negroid, Khoisanoid, Caucasoid, Australoid, and Melanesian peoples. Generally, it appears that the first molars of Negroid, Khoisanoid, Caucasoid, and Australoid peoples show a relatively low incidence of accessory distal cuspulids, the Melanesians tend to have a somewhat higher incidence of this feature, while the tuberculum sextum appears to be present most commonly on the first molars of Mongoloid populations. The same interpopulational differences tend to exist with regard to the incidence of this cuspulid on the second permanent molars, the most notable exception being the Australoid peoples, who have this feature relatively frequently on this tooth. With regard to the third mandibular molar, the Caucasoid populations tend to exhibit a tuberculum sextum less frequently than most others and, again, the incidence of this cuspulid on the M, is greater in Mongoloid populations than in others. It would appear that there is no consistent relationship between the fre- quency of accessory distal cuspulids on the first, second and third molars within populations. That is, a tuberculum sextum appears to occur in declining order of frequency on M,-M,-Msg respectively about as often as on M,—-M,—-M, respectively (Table 19). However, in all population studies in which all three molar types were examined, the third molar was found to possess a tuberculum sextum more commonly than the second molar. Although the third molar shows a higher frequency of five cusps than does the second molar, this can hardly be related to the higher incidences of tubercula sextum in the third molar, as the first molar, which shows the highest frequency of five cusps, has a tuberculum sextum in a higher percentage of cases than the M, in about the same number of populations in which the second molar shows this cuspulid more frequently than the M, (Table 19). The South African Negro and San both show the tuberculum sextum most frequently on the third molar, followed in descending order by the first and second molars respectively. Of the other three Negro populations examined PERMANENT MOLARS OF NEGRO AND SAN 193 (Table 19), two (west African and American) have molar frequency orders of the tuberculum sextum of M;-M,—M,, while the third (east African) has a molar frequency order of M,-M,—Mg. The frequencies of the tuberculum intermedium recorded for the South African Negro and Kalahari San are given in Table 20. The Negro female shows this accessory lingual cuspulid more frequently than the Negro male on all three molars. The San female tends to have a tuberculum intermedium more often than the San male on the first molar but less frequently than the male on the third molar. However, the sexual differences in both populations are slight and not statistically significant. TABLE 20 Incidence of tuberculum intermedium on lower permanent molars of the South African Negro and Kalahari San. M, M, M; Population Sex N yf N Ss N Ys Negro Males ase hie 32, tsa 4 ies 30 ~=10,0 Remale, 42°95 ~: iS 1728 2 1,9 2" 21356 Combined — - .. S0 1455 6 128) 42 10,8 San Males 2 so .. ee 7 0 0,0 3 7,9 Female 2. = . he; 2255 0 0,0 De 4,9 GCombmed © =" = 2 205 O.=-0:0 5 6,3 The South African Negro shows a tuberculum intermedium less frequently on the first molar but more frequently on the third molar than does the San. Only 1,5 per cent of Negro second molars were found to show this cuspulid, while no San M, evinced a tuberculum intermedium. The frequency differences on all molars amongst these two populations are slight and not statistically significant. Jacobson (1967) recorded the presence of the tuberculum intermedium for the South African Negro and, although he gave no indication of the criteria used in the recording of this feature, his data are similar to those reported here. Jacobson’s frequencies are, however, consistently (albeit slightly) larger than those given here. These discrepancies are probably due to the fact that different classificatory standards may have been employed, and also that Jacobson (1967) counted teeth rather than individuals. In her study of Khoisan teeth, Kiernberger (1955) recorded the frequencies of accessory cusps on the lower molars, but did not differentiate between the tuberculum sextum and tuberculum intermedium. Rather, she regarded the presence of either one or both on a molar as constituting a single crown category. Numerous workers have, like Kiernberger (1955), recorded the presence of additional cuspulids and have not separated the tuberculum intermedium and the tuberculum sextum (Hrdli¢ka 1910; Janzer 1927; Rosenzweig & Zilber- man 1967, 1969; Sofaer, MacLean & Bailit 1972; Sofaer et al. 1972; Berry 1976). Frequencies of the tuberculum intermedium recorded for the permanent lower molars of various human populations are tabulated in Table 21. A 194 ANNALS OF THE SOUTH AFRICAN MUSEUM TABLE 21 Percentage frequencies of tuberculum intermedium on permanent lower molars reported for various human populations. Sexes combined. Group Population M, M, M; Reference Mongoloid Amerindian: 4... 6,6 0,0 2,0 Nelson 1938 0,0 0,0 6,0 Hellman 1928 Eskimoid . .. . 0,0 0,0 0,0 Hellman 1928 0,0 0,7 0,0 Pedersen 1949 Chinese Se Ae 0,0 5,0 0,0 Hellman 1928 12,0 8,0 De Terra 1905 11,8 Turner 1976 Japanese... 6,7 Hanihara 1970 0,3 Suzuki & Sakai 1957 sil 0,2 Suzuki & Sakai 1956c 6,7 Turner 1976 1a) Turner 1976 58,0 14,0 DAES Turner 1979 ARI oP ie ee 4,3 Hanihara 1970 0,0 Suzuki & Sakai 1957 4,3 Turner 1976 4,3 0,0 2,9 Turner & Hanihara 1977 Polynesian. . . 23,0 14,0 17,0 Turner & Scott 1976 7,0 1,1 Suzuki & Sakai 1973 Hawaiian... 15,0 7,0 3,0 Katich & Turner 1975 Yapese..ar 31,0 12,0 20,0 Harris et al. 1975 Australoid Austral. Aborigine 0,0 0,0 0,0 Hellman 1928 6,3 De Terra 1905 Melanesian Melanesian bis 9,0 5,0 9,0 Dahlberg 1961 Papuam ss cy. 20,0 De Terra 1905 New Britain... 10,9 0,0 8,8 Turner & Swindler 1978 Caucasoid European. Les) 1,0 2,6 De Terra 1905 155 Ded Axelsson & Kirveskari 1979 American . . . 3,0 2,0 0,0 Hellman 1928 Negroid West African . . 2,0 5,0 6,0 Hellman 1928 South African. . 14,5 1S 10,8 Present study American... 22,0 4,0 13,0 Hellman 1928 Khoisanoid Soi ee Sie 20,3 0,0 6,3 Present study Note that the percentage frequencies recorded for the San and South African Negro are not strictly comparable with the results of a number of these studies because different definitions of cuspal presence were used. number of workers whose data are presented in Table 21 regarded a post- metaconulid, an entoconulid and/or a large, separate median lingual cuspulid as a tuberculum intermedium. Had the present author also subscribed to the recognition of all three these features, instead of only the third, as a tuber- culum intermedium it is certain that the percentage frequencies would have been higher than those presented in Table 20. Even so, the incidences of the cuspulid on the first molars of the San and South African Negro are compara- tively high. Of the twenty-nine populations in which first molars have been examined for this feature (Table 21), six have higher frequencies than the South African Negro (Japanese of Turner 1979; Polynesians of Turner & Scott 1976; Hawaiians of Katich & Turner 1975; Yapese of Harris et al. 1975; Icelandic Caucasoids of Axelsson & Kirveskari 1979; American Negroes of Hellman PERMANENT MOLARS OF NEGRO AND SAN 195 1928). Of these populations, only four have a greater percentage incidence than the San (Japanese of Turner 1979; Polynesians of Turner & Scott 1976; Yapese of Harris et al. 1975; American Negroes of Hellman 1928). It should be noted that there are rather dramatic differences amongst the frequencies of M, cuspulids reported for Jomon Japanese by Turner (1979) and other Japanese populations, between the frequencies reported for Polynesians by Turner & Scott (1976) and Suzuki & Sakai (1973), and between the Icelandic Caucasoids (Axelsson & Kirveskari 1979) and other European and American Caucasoid populations. It is possible that the frequencies recorded by Turner (1979), Turner & Scott (1976), Harris et al. (1975), and Axelsson & Kirveskari (1979) are inflated because of the classificatory criteria employed by them. The incidences and frequencies of the presence of a fovea anterior on the lower molars of the San and South African Negro are presented in Table 22. TABLE 22 Incidence of fovea anterior on lower permanent molars of the South African Negro and Kalahari San. M, M, M; Population Sex N vps N Wa N vA Negro Males “5 >. 148 69,8 oy Sikes 144 50,0 Female’ =< ... 62 69,7 69 67,6 D4 6355 Gombined ~- . ~210 (69'8 236076054 198 53,1 San Male = 2 SF. oy ON? 41 82,0 12° “35,3 nemale aan 43 76,8 Po OA 123553 Combined... 95 84,1 68). 72,3 DAY 3553 The first molars of the Negro male and female show this feature in an almost identical percentage of cases and, while the female possesses a fovea anterior on the second molar more commonly than does the male, this difference is statistically insignificant. The frequency of this fossid on the third molar is significantly greater in the Negro female (X? = 4,82; p < 0,05) than in the male. On the other hand, the fova anterior is present on the first and second molars significantly more frequently (X? = 4,40; p < 0,05 and X? = 4,98; p < 0,05 respectively) in the San male than in the San female. While a signifi- cant degree of sexual dimorphism in the presence of this fossid is apparent in the third molars of the Negro, both sexes of the San show the same frequency of the fovea anterior on the wisdom teeth. In all instances where significant sexual dimorphism was apparent, the differences were barely significant. The fovea anterior and cusp number were found to be independent variables on all three molars of the San, on the first molars of the Negro and on the third molars of the Negro female. These two features were found to be dependent on the second molars of the Negro male (X? = 4,19; p = 0,04) and the Negro female (X? = 7,92; p = 0,005) and on the third molars of the Negro male (X? = 12,22; p = 0,002). In these latter three instances the fovea anterior tends to be present on five cusped teeth and absent on molars with only four cusps. 196 ANNALS OF THE SOUTH AFRICAN MUSEUM The San male possesses a fovea anterior significantly more frequently than the Negro male on both the first (X? = 10,80; p < 0,005) and second (X? = 10,54; p < 0,005) molars. The San male shows this fossid less frequently than the Negro male on the third molar, but this difference is not statistically significant. The Negro female possesses a fovea anterior less commonly than the San female on the first molar but more frequently on the second molar; these differences are slight, however, and are statistically insignificant. The Negro female displays this fossid more frequently than the San female on the third molar (X? = 7,84; p < 0,01). Notwithstanding the sexual dimorphism apparent in the expression of the trigonid basin in both the San and Negro, the data were pooled for purposes of overall comparison (Table 22). The San displays a fovea anterior signifi- cantly more frequently than does the Negro on both the first (X? = 8,67; p < 0,005) and second (X? = 4,65; p < 0,05) mandibular molars. Conversely, the Negro possesses this mesial fossid significantly more frequently than the San on the third molar (X? = 7,28; p < 0,01). In both populations, and in both sexes of these populations the first molar shows the highest frequency of mesial fossids, following in decreasing order of commonality by the second and third molars respectively. The incidences and frequencies of the fovea posterior on the mandibular molars of the South African Negro and Kalahari San are presented in Table 23. No significant sexual dimorphism in the presentation of this character on any of the molars was found within either the San or Negro samples. TABLE 23 Incidence of fovea posterior on lower permanent molars of the South African Negro and Kalahari San. M, M, M; Population Sex N We N Yes N A Negro Maller nae 29 Bi N27 47 16,3 emalens eae 19 21,3 15 14,3 My 17/4! Combined . . AS males 5206 sl3ell 62 16,6 San Maes eee es ee 11 20,4 3 Tol 0 0,0 inemales ae 10 19,2 4 10,0 1s Combined. . Die lO8S le 4285 1 1) The San male shows the distal fossid more frequently than the Negro male on the first molar, while the Negro male possesses a fovea posterior more commonly on the second molar; these differences are slight and not statistically significant. The Negro male displays a fovea posterior significantly more frequently than does the San male on the third molar (X? = 5,56; p < 0,025). The San female displays this distal fossid less frequently than does the Negro female on all three molars; the differences with regard to the first and second molars are slight and statistically insignificant. The Negro female fre- quency is significantly higher on the third molar (X? = 4,79; p < 0,05). PERMANENT MOLARS OF NEGRO AND SAN 197 Comparison of the sexually pooled data (Table 23) indicates that, while the San tends to show a fovea posterior more commonly than the Negro on the first molar, the Negro has a higher frequency of this fossid on the second. These differences are slight and not statistically significant. The Negro possesses a significantly higher frequency of the fovea posterior than does the San on the third molar (X? = 10,19; p < 0,005). Chi-square evaluation of the fovea posterior and cusp number revealed these two features to be independent on all three molars of both sexes of the San and Negro. It is apparent that the fovea anterior is considerably more common than the fovea posterior on all lower molars of the San and Negro (cf. Tables 22-23). Biggerstaff (1968) recommended that the combination of the presence or absence of the fovea posterior and cuspal number was a more useful criterion for the classification of lower molars than the traditional occlusal patterns. In a study of dental casts of 199 pairs of like-sexed twins he (Biggerstaff 1975) found that males had a significantly higher frequency of molars with five cusps and fovea posterior (his ‘5fd’ type) than did females. In the present samples of Negro and San molars, however, females possess a higher frequency of ‘5fd’ molars, except for the Negro Mg, and the San Mg, in which instances the female frequencies are slightly lower than those exhibited by the male. It has been shown that cusp number and the fovea posterior are independent variables (at least on the Negro and San molars). In addition, both sexes of the San and Negro exhibited four-cusped molars with the fovea posterior. Thus, Bigger- staff’s (1968, 1975) method of crown classification appears to be incomplete, and even if complete it would serve no more useful purpose than the classi- fication of molars according to the ‘traditional’ cusp-groove pattern relationships. It has been claimed that ‘the tuberculum sextum probably appears most frequently on five-cusped teeth having a distal fovea and distal marginal ridge’ (Biggerstaff 1968: 444). Quite the contrary is evident in the Negro and San lower molars. In the majority of cases the tuberculum sextum and the fovea posterior do not occur together on the same tooth, probably because a tuber- culum sextum (especially a large cuspulid) obliterates the distal fossid by filling up the distolingual corner of the crown (Robinson 1956). The presence of crenulate or wrinkled enamel was recorded here for the lower molars of the South African Negro and San (Table 24). In the Negro, the female shows a higher frequency of crenulate enamel than does the male on all three molars. The degree of sexual dimorphism in enamel appearance on the first and second molars is slight and the differences are not statistically significant. In the third molar, however, the frequency of crenulate enamel in females is significantly greater than that in males Ce NOS: pp =< 0,005). The San male shows crenulate enamel more commonly than the San female on the first molar, but on the second and third molars the female tends to 198 ANNALS OF THE SOUTH AFRICAN MUSEUM TABLE 24 Incidence of crenulate enamel on lower permanent molars of the South African Negro and Kalahari San. M, M, M; Population Sex N WA N vA N Yep Negro Mae He - oa 11 4,5 13 4,3 60 19,9 Female 7 6,9 8 Tes) 32 36,4 Combined... 18 54 21 5,1 92 23,6 San Wale a San ours yy 3,3 1 2,1 3 8,6 Hemale 1 1,8 1 Deh 3 9,4 Combined... 3 2,6 Z, 2,4 6 §869,0 possess crenulate enamel more frequently. The sexual differences amongst the lower molars of the San are slight and statistically insignificant. The Negro male tends to possess wrinkled enamel more frequently than the San male on all three mandibular molars, but in no instance was the differ- ence in frequency between them found to be statistically significant. The Negro female shows crenulate enamel more commonly than the San female on all molars, and while the differences in frequency on the first and second molars are insignificant, in the third lower molar the Negro female incidence is sig- nificantly greater than that of the San female (X? = 8,27; p < 0,005). Thus, the Negro tends to possess crenulate enamel more commonly than does the San on all three permanent mandibular molars. The Negro frequencies are only slightly higher than those of the San for the first and second molars, and in these instances the sexually pooled population differences are not statistically significant. The third molar of the Negro is crenulate in appearance con- siderably more often than is this tooth in the San, and here the differences in frequencies are statistically significant (X? = 7,23; p < 0,01). Crenulate enamel on the lower molars of both the San and Negro appears to be more common on the third molar than on either of the mesial two molars. In both populations the expression of crenulate enamel by the first and second molars is nearly of the same frequency. Due to the inherent difficulty of objectively classifying crenulate enamel, the number of observations of this feature recorded for recent human teeth is understandably low. Jacob (1967) stated that he believed enamel wrinkling to be a characteristic feature of Mongoloid dentitions. Kiernberger (1955: 31) maintained that ‘das wichtigste Charakteristikum des Khoisanidenzahnes ist die von Adloff festgestellte Schmelzrunzelung der Molarenkrone’. The fre- quencies recorded here for the San molars certainly do not support Kiernberger’s statement; rather, they imply quite the contrary. Pedersen (1949: 103) recorded that ‘pronounced wrinkles on the occlusal surfaces were seen in seven out of twenty lower third molars in the skulls [a fre- quency of 35,0 per cent—cf. Negro female incidence] and in many cases of the living Eskimo’. He noted also that these wrinkles were absent on all first and second molars examined by him, but that the wear shown by these teeth had probably eradicated any trace of enamel wrinkles. PERMANENT MOLARS OF NEGRO AND SAN 199 It is evident that smooth occlusal enamel is characteristic of the lower molars of the San, and of the first and second molars of the South African Negro. The third molar of the Negro (especially of the female) tends to show a higher incidence of secondary and tertiary enamel cristids, but even here the incidence cannot be considered to be high. DISCUSSION SEXUAL DIMORPHISM Significant sexual dimorphism in the morphology of the permanent man- dibular molars is relatively low in both the San and the South African Negro. Such dimorphism is expressed in only three of the eight features examined: the deflecting wrinkle, the fovea anterior, and the presence of crenulate enamel (Table 25). The incidences of metaconid deflecting wrinkles were significantly dimorphic on only the first molar of the Negro, while the incidences of crenulate enamel were found to be significantly different on only the third molar of the Negro. Significant sexual dimorphism is expressed most commonly with regard to the fovea anterior. The sexual incidences of this feature are significantly different on the first and second molars of the San and on the third molars of the Negro. Of the five instances of significant sexual dimorphism found in the Negro and San molars (Table 25) the sexual differences were barely signifi- cant in four. A highly significant degree of sexual dimorphism is shown only by the incidence of crenulate enamel on the Negro third molars. However, crenulate enamel is very difficult to classify objectively ; enamel wrinkles may be obliterated by slight wear, and the genetic factors influencing the expression of these wrinkles may be easily overshadowed by environmental factors such as dietary fluorine intake, disease, and nutritional status. TABLE 25 Summary of morphological features in which significant sexual dimorphism was demonstrated in the South African Negro and Kalahari San. M, M, M; Character Negro San Negro San Negro San 1 soe) ee LER ieee a 2 ae ae SOS ghee re es 3 =a ea ag eee yeas 4 Sey eee cay gies eee 5 b — _— —_—- — 6 _— b — pb b — 7 sla A cae =S ety ae 8 —_—- — —- — a a—significant difference at p < 0,005; b—significant difference at p < 0,05. Characters: 1—principal cusp number; 2—tuberculum sextum; 3—tuberculum intermedium; 4—primary occlusal pattern; 5—deflecting wrinkle; 6—fovea anterior; 7—fovea posterior; 8—crenulate enamel. 200 ANNALS OF THE SOUTH AFRICAN MUSEUM Thus, there is a rather low incidence and degree of sexual dimorphism in the features examined here on the lower permanent molars of the South African Negro and Kalahari San. The dimorphism shown by these populations is comparable to, or perhaps slightly less than, that described for the permanent mandibular molars of other population groups. De Villiers (1958) examined the expression of sexual dimorphism in some sixty-seven non-metrical features of the South African Negro skull. She found that significant sexual differences are manifest in only fourteen of these features, and that sexual dimorphism is largely associated with the face and jaws and it is most pronounced in man- dibular characters. De Villiers (1968) concluded that the amount of sexual dimorphism displayed in the skull of the South African Negro is comparable to, or slightly less than, that shown by other groups. CHARACTERISTICS OF SAN AND NEGRO MOLARS While both the incidence and degree of sexual dimorphism within the San and Negro molars appear to be rather low, the significant differences in the morphological characters of the three mandibular molars amongst these two populations are more common (Table 26). Thus, seven of eight morpho- logical features examined show significant interpopulational differences variously on one or more of the permanent molars. Two features (fovea posterior and crenulate enamel) have significant interpopulational differences on the third molar only, one feature (fovea anterior) shows significant San—Negro differences on all three molars, one character (tuberculum sextum) is significantly different on the first and second molars, and some three non-metrical features (principal cusp number, primary occlusal pattern, and deflecting wrinkle) show significant interpopulational differences on only the second molar. Of the ten instances of significant interpopulational differences encountered, only two (deflecting wrinkle and fovea anterior on the M,) were found to be barely significant statistically (Table 26). In the first molar only two of eight features show significant differences of incidence between the San and Negro, three features on the third molar display such differences while five characters show significant populational differences on the second molar. Thus, the fovea anterior and the tuberculum sextum appear to be respectively the first and second most useful discriminatory characters of the mandibular molars, while the second molar seems to display the greatest number of differences between the San and South African Negro populations (Table 26). On the basis of the foregoing analyses a general characterization of the morphology of the permanent mandibular molars of the San and South African Negro may be summarized. In view of the relatively low degree of sexual dimorphism apparent in the features examined, the characterizations presented here refer to sexually pooled population samples and sexual differences are enumerated only in those instances where they proved to be statistically signifi- PERMANENT MOLARS OF NEGRO AND SAN 201 TABLE 26 Summary of morphological features in which significant (sexually pooled) population differences amongst the South African Negro and Kalahari San were demonstrated. Character M, M, M; 1 a — DD a a oo 3 aus Lhe as. 4 — a _ 5 == c — 6 a c b 7 — a 8 — — b a—significant difference at p < 0,005; b—significant difference at p < 0,01; c—significant difference at p < 0,05. Characters: 1—principal cusp number; 2—tuberculum sextum; 3—tuberculum inter- medium; 4—primary occlusal pattern; 5—deflecting wrinkle; 6—fovea anterior; 7—fovea posterior; 8—crenulate enamel. cant. Lest these characterizations of lower molar morphology be viewed as typological, the frequencies of each trait are alluded to by the use of terms such as ‘almost always’, ‘usually’ and ‘generally’. Arbitrarily defined frequency limits have been assigned to these terms as they are used here; thus, ‘almost always’ indicates a frequency between 90-100 per cent, ‘usually’ indicates a frequency between 80-90 per cent, and ‘generally’ refers to an incidence of between 70-80 per cent. FIRST PERMANENT MOLAR Negro: almost always five-cusped; tuberculum sextum usually to almost always absent; tuberculum intermedium usually absent; primary occlusal pattern generally to usually in the form of a Y; metaconid cristid generally either straight and strongly or straight and weakly developed in male and commonly strongly developed in female with deflecting wrinkle usually absent in the male and generally absent in the female; fovea anterior generally present; fovea posterior usually absent; occlusal surface enamel usually to almost always smooth. San: almost always five-cusped; tuberculum sextum usually absent but present significantly more often than in the Negro; tuberculum intermedium generally or usually absent; primary occlusal pattern usually in the form of a Y; meta- conid cristid generally either straight and strongly or straight and weakly developed with deflecting wrinkle generally to usually absent; fovea anterior almost always present in the male and generally present in the female and present significantly more often than in the Negro; fovea posterior generally to usually absent; occlusal surface enamel almost always smooth. 202 ANNALS OF THE SOUTH AFRICAN MUSEUM SECOND PERMANENT MOLAR Negro: almost equally five or four-cusped; tuberculum sextum almost always absent; tuberculum intermedium almost always absent; primary occlusal pattern more commonly in the form of a +; metaconid cristid usually either weakly developed or absent with a deflecting wrinkle almost always absent; fovea anterior somewhat more commonly present than absent; fovea posterior usually absent; occlusal enamel usually to almost always smooth. San; generally five-cusped, and significantly more frequently than in the Negro; tuberculum sextum usually absent but present significantly more frequently than in the Negro; tuberculum intermedium almost always absent; primary occlusal pattern almost equally in the form of either a Y or +, but significantly more frequently in the form of a Y than in the Negro; metaconid cristid generally straight and weakly developed and deflecting wrinkle usually to almost always absent, and with a significantly higher frequency of weak crests and a signifi- cantly lower incidence of deflecting wrinkles than in the Negro; fovea anterior usually present in the male and only somewhat more commonly present than absent in the female, present significantly more frequently than in the Negro; fovea posterior usually absent; occlusal enamel almost always smooth. THIRD PERMANENT MOLAR Negro; generally to usually five-cusped; tuberculum sextum generally to usually absent; tuberculum intermedium usuall absent; primary occlusal pattern usually in the form of either a + or an X, with almost equal frequencies of both; metaconid cristid almost always either weakly developed or absent, with almost equal frequencies of both expressions; fovea anterior either present or absent with almost equal frequency in the male and somewhat more com- monly present in the female; fovea posterior usually absent; occlusal enamel generally smooth. San: generally to usually five-cusped; tuberculum sextum more commonly to generally absent; tuberculum intermedium usually absent; primary occlusal pattern generally either in the form of a + or an X with almost equal frequen- cies of both and, as in the Negro, with only a slightly lower incidence of the Y pattern; metaconid cristid usually either weakly developed or absent and, as in the Negro, a deflecting wrinkle is usually to almost always absent; fovea anterior more commonly absent than present and present significantly less frequently than in the Negro; occlusal enamel usually smooth, and crenulate in appearance significantly less frequently than in the Negro. OCCLUSAL MORPHOLOGICAL INTERRELATIONSHIPS Almost all of the features examined here show the same intermolar inci- dence relationships (e.g. a given character is most commonly expressed on the M,, less frequently shown by the M, and least commonly evinced by the Ms) in both the San and Negro. In both populations the five principal cusps, the PERMANENT MOLARS OF NEGRO AND SAN 203 metaconid deflecting wrinkle and the tuberculum intermedium are most com- monly shown by the first molar, followed in decreasing order of frequency by the third and second molars respectively (i.e. frequencies M, > M, > M.,). The fovea anterior and the Y pattern are expressed most frequently by the first molar, followed in decreasing incidences by the second and third molars respectively (i.e. M, > M, > Ms). Conversely, the tuberculum sextum and the X pattern are shown most commonly by the third molar, followed in decreasing order of frequency by the second and first molars _ respectively (i.e. M; > M, > M,). In both the San and the Negro the H pattern is evinced only on third molars. Crenulate enamel is displayed most frequently on this tooth with considerably lower, and almost equal, frequencies shown by the first and second molars. The + pattern is shown most frequently by the second molars, while the third and first molars show decreasing frequencies of this pattern (i.e. M, > Ms; > M,). With respect to the fovea posterior, however, this character shows a frequency relationship in the South African Negro of M, > M, > M, and in the San of M, > M, > Ms. On all three molars of both the Negro and the San the fovea anterior is present more often than the fovea posterior. Also, in both populations, while the tuberculum intermedium is shown more frequently than the tuberculum sextum on the first molar, on the second and third molars the tuberculum sextum is present considerably more frequently than the tuberculum intermedium. Paired chi-square evaluations of interdependence have been performed for most of the characters examined. In these evaluations either no inter- dependence or sometimes barely significant levels of dependence were found. Thus, in both the San and the South African Negro the principal cusp number and primary groove pattern appear to have a sometimes low level of dependence, but generally these two characters seem to be independent. The primary occlusal pattern appears to be independent of metaconid cristid morphology, and in particular, the Y pattern and the deflecting wrinkle are not dependent. The presence of neither the fovea posterior nor the fovea anterior is dependent upon principal cusp number. Similarly, the presence of crenulate occlusal enamel is independent of principal cuspal number on the lower molars of the San and South African Negro. Thus, until the type of genetic control over the expression of lower molar dental variants has been established, it would seem that each of these features should be categorized separately. That is, classificatory schemes that combine two or more characters (e.g. Y-5, +-4, X-5, 5fd, etc.) should not be used in odontological studies. BIOLOGICAL INTERRELATIONSHIPS OF THE SOUTH AFRICAN NEGRO AND THE KALAHARI SAN The term ‘Koisan’ (or, as it is usually spelt, Khoisan) was proposed by Schultze-Jena (1928: 211) ‘als gemeinsamer Rassenname fiir Hottentotten und Buschmanner’. The name Khoisan is most commonly used today in a generic 204 ANNALS OF THE SOUTH AFRICAN MUSEUM sense to include the Khoi (Hottentots) and the San (Bushmen). The biological interrelationship of the Khoisan (and especially the San) and African Negroes have been the subject of considerable study. Until about the middle of the present century, most workers were concerned with producing a classification or taxonomy of African peoples, and these workers generally expounded a typological approach to the problems of interrelationship definition. No useful purpose would be served here by reviewing these past polemics. More recently, studies of the Khoisan and Negro peoples have been based upon an examination of the limits of phenotypic variability of gross morphological features in these populations. Amongst the constellation of features thus examined a few of the more notable that serve to characterize the San are: mesometriocranial and pentaganoid cranium; mesoproscopic face and euryene upper face; ortho- gnathism; mesochonch orbits; brachystaphaline palate; small postglenoid tubercles; faint glabella and weak supraciliary ridges; nasal bones highly obtusely angulated (De Villiers 1968); frontal processes of the maxillae highly obtusely angulated (Grine 1979); skin showing a lower mean reflectance and therefore lighter in colour than that of the Negro (Weiner et al. 1964); head hair tending to consist of tightly spiralled tufts; small, commonly lobeless ears with overrolled helices; relatively small stature (Tobias 1955-6, 1960, 1966: Singer & Weiner 1963); steatomeria and steatopygia (De Villiers 1961; Krut & Singer 1963); marked and nearly universal macronymphia (De Villiers 1961); ithyphally (Drury & Drennan 1926); fewer dermatoglyphic whorls and more arches; and high frequency of palmer patterns in areas II-IV (Tobias 1961). Tobias (1972), Nurse & Jenkins (1977), and Singer (1978) have compiled somewhat more extensive lists of gross morphological features that tend to characterize the San and delineate them from other African groups. Most, if not all, of the foregoing characters are probably polygenic and have complex modes of inheritance. In addition, the relationship between genotype and phenotype in not a few of these characters is made more complex by the modi- fying influences of the environment. Consequently, the assessment of these characters as genetic markers is extremely difficult. Since the sixties the Khoisan and Negro peoples of southern Africa have been the subject of a number of genetic studies (Tobias 1966, 1972; Nurse & Jenkins 1977). The essential genetic unity of sub-Saharan Africa has been shown by these studies. More particularly, the strong genetic affinities of the South African Negro with the Negro populations of west and central Africa have been stressed (Tobias 1972). Although these genetic studies have demon- strated that the Khoisan peoples have more in common genetically with Negroes than either group has with non-African peoples, a number of differences between the San and Negro have been detected. The Abantu phenotype was first characterized by Brain (1966) and thus named because he believed it to be characteristic of the South African Bantu- speaking Negro. Jenkins (1974) has shown this allele to be much commoner in the San (and Khoi) than in the southern African Negro and, because of its PERMANENT MOLARS OF NEGRO AND SAN 205 virtual absence in Negroes to the north, he has postulated that the 42" red cell antigen is primarily a Khoisan character that has been acquired secon- darily by Negro peoples in their southward migrations. Similarly, the Duffy red cell antigen, Fy, occurs with high frequency in the Khoisan, while Negroes often lack any Duffy antigen (Nurse & Jenkins 1977). Nurse & Jenkins (1977) have proposed that the Fy* gene has been acquired by the southern African Negro from the Khoisan. Jenkins & Corfield (1972) have claimed that the acid phosphatase allele, P’, is almost certainly a Khoisan marker, and it has been proposed that this gene was similarly acquired by the Negroes in the course of their southward migrations (Nurse & Jenkins 1977). However, Santachiarra- Benerecetti et al. (1977) and Ojkutu et al. (1977) reported P! frequencies of 0,17 and 0,05 in the Babinga Pygmies and the Yorba of Nigeria respectively, and, as such, this allele may not be a very useful marker in this context. Transferrin D,, a f-globulin variant, seems to represent an essentially San character in southern Africa, and its presence in southern African Negroes has been ascribed to San admixture (Nurse & Jenkins 1977). In a survey of the distribution of variants of the 6-phosphogluconate dehydrogenase (6PGD) red cell enzyme in southern Africa, Jenkins & Nurse (1974) found that the PG D¢ allele is not often present in San populations. They proposed that its presence in the San is an indication of Negro admixture. Glucose-6-phosphate dehydrogenase (G6PD) deficiency has been shown to be much less common in the San than in Negro populations (Charlton & Bothwell 1961). However, both of the characteristically Negro variants have been identi- fied in the San, and this has been ascribed to Negro admixture (Nurse & Jenkins OTT): The Gm system of serum protein polymorphisms appears to be one of the most useful so far discovered for the elucidation of anthropological prob- lems in southern Africa (Nurse & Jenkins 1977). The Gm>-* haplotype appears to be characteristic of the Negro, and the Gm!'8 (and to a lesser degree the Gm?!) haplotype is common amongst the San peoples, while Gm1-® is rare and Gm?! is virtually absent in other sub-Saharan African populations (Jenkins & Steinberg 1966; Jenkins et al. 1970; Steinberg et al. 1975). Only a few of the apparent genetic differences found so far among the San and Negro are mentioned above (see Tobias 1972; Nurse & Jenkins 1977). It appears that while some alleles are characteristically San (or Khoisan) (e.g. Abantu, acid phosphatase P', Fy®, Gm'-!8) others are characteristic of the Negro (e.g. peptidase Pep*?, Gm!->»°). Furthermore, the frequency distributions of these various genes amongst the San and Negro populations seem to indicate a certain amount of genetic exchange between the two groups. The quantity of genetic exchange is difficult to ascertain because of uncertainties about the part played by drift or selection in the attainment of the frequencies of certain alleles in the San and Negro. For example, the transferrin D, variant, which is frequently found amongst the San and less commonly shown by the southern African Negro, has been shown to be electrophoretically and chemically 206 ANNALS OF THE SOUTH AFRICAN MUSEUM identical to that found among Australian aborigines (Krik et al. 1964; Wang et al. 1967). The San also manifest, in common with the Australian aborigines, low or absent frequencies of S in the MNSs Henshaw system and the occasional presence of adults with the Lewis Le (a+b-+) phenotype (Nurse & Jenkins 1977). These similarities may reflect parallel adaptations to the selective pressures imposed by similar environments and modes of life. How do the results obtained from the present study of the occlusal mor- phology of the permanent lower molars of the San and the South African Negro compare with the results of other morphological and genetic studies on these people? Comparative data on the principal cusp numbers, the primary occlusal patterns, the tuberculum sextum and tuberculum intermedium for other African -and New World Negro populations have been recorded (see Tables 4—6, 8-14, 19, 21). Data for the frequencies of the tuberculum sextum and tuberculum intermedium are available for relatively few non-South African Negro popu- lations. The South African Negro frequencies for the tuberculum sextum on all three molars fall within the frequency ranges for other Negro populations (Table 19) and the South African Negro frequencies for the tuberculum inter- medium on the first and third molars fall between the frequencies shown by the west African and American Negro populations. The South African Negro very rarely shows a tuberculum intermedium on the second molar; this frequency is lower than that of any other Negro population and closely approximates the San incidence (Table 21). Considerably more comparative Negro data are available for principal cusp number and primary occlusal pattern frequencies. The frequency of five-cusped first mandibular molars in the South African Negro is amongst the highest recorded for Negro populations and closely approximates the San frequency (Table 4). Although the incidence of five-cusped second per- manent molars in the South African Negro (47,6 per cent) is significantly lower than the San incidence (78,4 per cent), frequency for the South African Negro is considerably higher than for any other Negro population so far examined (Table 5). It is possible that the relatively high frequency of five-cusped second molars in the South African Negro is due to a certain amount of San admixture. Similarly, the San and South African Negro show comparatively high frequen- cies of five-cusped third mandibular molars. Only Hellman’s (1928) west African population shows an incidence comparable with that of the South African Negro (Table 6). On the other hand, the South African Negro possesses the lowest frequency of the Y pattern on the first molar of any Negro population so far studied (Tables 9-10). The San shows a higher (albeit insignificantly so) incidence of this pattern than the South African Negro on the first mandibular molar, and the San incidence is thus closer to those shown by non-South African Negroes than it is to the South African Negro frequency. Similarly, the South African Negro shows a significantly lower frequency of the Y pattern on the second PERMANENT MOLARS OF NEGRO AND SAN 207 mandibular molar than does the San, and the incidences of almost all other Negro populations are higher than that of the South African Negro and thus closer to the San frequency (Tables 11-12). Thus, in certain features of the lower molars (e.g. principal cusp number, and perhaps, the tuberculum intermedium on the M,) the San and South African Negro seem to display frequencies that are more similar to each other than either is to most non-southern African Negroes. And this is evident despite the fact that in some of these instances (e.g. principal cusp number on the M,) the San and South African Negro frequencies are significantly different. In other characters (e.g. the primary occlusal pattern), however, the South African Negro frequencies appear to be further removed from those of the San than the incidences shown by other Negro populations. As discussed previously for each character, the San and other African (Negro) populations generally show frequencies more in common compared with the differences that separate Subsaharan Africans from most other racial constellations. Thus, the results of the present study on the mandibular molar morphology of the San and South African Negro are compatible with, and to some degree corroborate, the conclusions reached by others who have examined gross morphological and gentic features of these peoples (e.g. De Villiers 1968; Rightmire 1970; Tobias 1972; Nurse & Jenkins 1977). Although there is evidence that at least some dental features are monogenic, the modification by environmental factors of the genetic control over the expression of these dental traits and the degree to which these variants are selectively affected by wear and disease are far from completely understood. Because of these problems, Berry (1976) has questioned the practical value of minor crown variants in anthropological studies. Nevertheless, when used in conjunction with the results of studies on other gross morphological features and the results of genetic studies on the people in question, dental traits seem to convey informa- tion as reliable as that gained from other, non-dental characters. Perhaps the most serious drawback to the anthropological usefulness of dental traits is the question of the usefulness, for comparative purposes, of data published by different workers in view of the inter-observer differences in the scoring of dental variants. SUMMARY A series of morphological characters of the permanent mandibular molars of the South African Negro and Kalahari San have been investigated. These features include the principal cusp number, the presence or absence of the tuberculum sextum and tuberculum intermedium, the form of the primary occlusal pattern, the presence or absence of a deflecting wrinkle and the mesial and distal foveae, and the appearance of the occlusal enamel. The definitions and various methods employed in the classification of these characters are reviewed. 208 ANNALS OF THE SOUTH AFRICAN MUSEUM Significant sexual dimorphism in the morphology of the mandibular molars is relatively low in both the Sand and Negro. Such dimorphism is expressed variously by these populations in only three of the eight features examined (i.e. the deflecting wrinkle, the fovea anterior and the presence of crenulate enamel), and even then sexual dimorphism is not expressed on all three molars for these three features. While there is a low incidence and degree of sexual dimorphism in the features examined, significant populational differences amongst the San and Negro are more common. The fovea anterior and the tuberculum sextum appear to be respectively the first and second most useful discriminatory characters, while the second molar displays the greatest number of significant differences between the San and the Negro. A characterization of the lower molar morphology of the San and Negro is provided. Paired chi-square evaluations of interdependence have been performed for most of the characters examined. Most of the features were found to be inde- pendent of one another. Principal cusp number and primary occlusal pattern appear to have a sometimes low level of dependence in the Negro, but in most instances these two characters seem to be independent. No relationship is evident between the presence of a deflecting wrinkle and the presence of a Y occlusal pattern. The biological differences and similarities of the San and Negro as deter- mined by both gross morphological and genetic studies are reviewed. A number of morphological and genetic features that may be considered characteristic of the San are found in varying frequencies in South African Negro populations. It is generally assumed that ‘hybridization’ has brought some of the ‘San alleles’ into the gene pool of the southern African Negro. Furthermore, it is postulated by most workers that these San characteristics were acquired secondarily by the Negro in the course of their southward migrations. In certain of the dental features examined here the San and South African Negro frequencies appear to be more similar to one another compared with frequencies shown by other Negro populations. Thus, the dental data presented here for the San and South African Negro are compatible with, and to some degree corroborate, what is known about the interrelationships of these peoples from non-dental morpho- logical and genetic data. Perhaps the most serious drawback to the anthropological usefulness of dental traits is the question of the comparative usefulness of data published by different workers in view of the inter-observer differences in scoring these traits. ACKNOWLEDGEMENTS The dental casts and associated field records of the San were made available to me through the generosity of Professor J. F. van Reenen. Permission to examine the material in the Raymond A. Dart Collection was granted by Professor P. V. Tobias. Most of the calculations were made using the I.B.M. PERMANENT MOLARS OF NEGRO AND SAN 209 360/175, Computer Centre, University of the Witwatersrand. I thank Professors T. Jenkins, P. V. Tobias and A. J. Perzigian for reading the various drafts of this manuscript and for their invaluable comments and advice. REFERENCES ABEL, W. 1931. Kritische Untersuchungen tiber Australopithecus africanus Dart. Gegenbaurs morph. Jb. 65: 539-640. Aptorr, P. 1937. Uber die primitiven und die sogenannten ‘pithekoiden’ Merkmale im Gebiss des recenten und fossilen Menschen und ihre Bedeutung. Z. Anat. Entwgesch., Leipzig 107: 68-82. ADLOFF, P. 1938. Das Gebiss von Sinanthropus pekinensis. Z. Morph. Anthrop. 37: 490-537. AXELSSON, G. & KIRVESKARI, P. 1977. The deflecting wrinkle on the teeth of Icelanders and the Mongoloid dental complex. Am. J. phys. Anthrop. 47: 321-324. AXELSSON, G. & KIRVESKARI, P. 1979. Sixth and seventh cusp on lower molar teeth of Ice- landers. Am. J. phys. Anthrop. 51: 79-82. BaliitT, H. L., Dewitt, S. J. & LEIGH, R. A. 1968. The size and morphology of the Nasioi dentition. Am. J. phys. Anthrop. 28: 271-288. Barnes, D. S. 1969. Tooth morphology and other aspects of the Teso dentition. Am. J. phys. Anthrop. 30: 183-194. BENNEJEANT, C. 1936. Anomalies et variations dentaires chez les Primates. Paris: Clermont- Ferrand. Berry, A. C. 1976. The anthropological value of minor variants of the dental crown. Am. J. phys. Anthrop. 45: 257-268. Beynon, A. D. 1971. The dentition of the Afghan Tajik. Jn: DAHLBERG, A. A. ed. Dental morphology and evolution: 271-282. Chicago: University of Chicago Press. BIGGERSTAFF, R. H. 1968. On the groove configuration of mandibular molars: the unreliability of the ‘Dryopithecus pattern’ and a new method for classifying mandibular molars. Am. J. phys. Anthrop. 29: 441-444. BIGGERSTAFF, R. H. 1975. Cusp size, sexual dimorphism, and heritability of cusp size in twins. Am. J. phys. Anthrop. 42: 127-140. BLEEK, D. F. 1927. The distribution of Bushman languages in South Africa. Festscrift Meinhof. Hamburg: L. Friederichsen. BRABANT, H. 1965. Observations sur l’evolution de la denture temporaire humaine en Europe occidentale. Bull. Grpmt. int. Rech. scient. Stomat. 8: 235-302. BRABANT, H. 1971. The human dentition during the Megalithic era. In: DAHLBERG, A. A. ed. Dental morphology and evolution: 283-297. Chicago: University of Chicago Press. BRABANT, H. & TWEISSELMANN, F. 1964. Observations sur l’evolution de la denture permanente humaine en Europe occidentale. Bull. Grpmt. int. Rech. scient. Stomat. 7: 11-84. BRAIN, P. 1966. Subgroups of A in the South African Bantu. Vox Sang. 11: 686-698. BREWER-CARIAS, C., LEBLANC, S. & NEEL, J. 1976. Genetic structure of a tribal population, the Yanomama Indians. XIII. Dental microdifferentiation. Am. J. phys. Anthrop. 44: 5-14. BROTHWELL, D. R. 1967. Some problems and objectives related to the study of dental varia- tion in human populations. J. dent. Res. 46: 938-941. CAIDEN, J, D, 1972. Dental variation in man. Jn: WASHBURN, S. & DOLHINOwW, R. C. eds. Perspectives on human evolution: 199-222. New York: Holt, Rinehart & Winston. CAMPBELL, T. D. 1925. Dentition and palate of Australian Aboriginal. Publs Keith Sheridan Fdn. Med. Res. 1: 1-123. CAMPUSANO, C., FIGUEROA, H., LAzo, B., PINTO-CISTERNAS, J. & SALINAS, C. 1972. Some dental traits of Diaguitas Indian skulls. Am. J. phys. Anthrop. 36: 139-142. CHAGULA, W. K. 1960. The cusps on the mandibular molars of East Africans. Am. J. phys. Anthrop. 18: 83-90. CHAPPEL, H. G. 1927. Jaws and teeth of ancient Hawaiians. Mem. B. P. Bishop Mus. 9: 1-20. CHARLTON, R. W. & BOTHWELL, T. H. 1961. Primaquine sensitive red cells in various races in southern Africa. Br. med. J. 41: 941-944. 210 ANNALS OF THE SOUTH AFRICAN MUSEUM CoRRUCCINI, R. S. 1972. The biological relationships of some prehistoric and historic Pueblo populations. Am. J. phys. Anthrop. 37: 373-388. DAHLBERG, A. A. 1949. The dentition of the American Indian. In: LAUGHLIN, W. S. ed. Papers on the physical anthropology of the American Indian: 138-176. New York: Viking Fund. DAHLBERG, A. A. 1961. Relationship of tooth size to cusp number and groove conformation of occlusal surface patterns of lower molar teeth. J. dent. Res. 40: 34-38. DAHLBERG, A. A. 1963. Analysis of the American Indian dentition. In: BROTHWELL, D. R. ed. Dental anthropology: 149-177. London: Pergamon. DE TERRA, M. 1905. Beitrdge zu einer Odontographie der Menschenrassen. Berlin: Berlinische Verlags. DE VivutErRs, H. 1961. The tablier and steatopygia in Kalahari Bushwomen. S. Afr. J. Sci. 57: 223-227. DE VILuiers, H. 1968. The skull of the South African Negro: a biometrical and morphological study. Johannesburg: Witwatersrand University Press. DeEvoTo, F. C. & CIGLIANO, E. 1970. Morphology of the permanent molars in pre-Columbian population of Santa Rosa de Tastil (Salta, Argentina). J. dent. Res. 49: 997. DevotTo, F. C. & PERROTTO, B. M. 1972. Groove pattern and cusp number of mandibular molars from Tastilian Indians. J. dent. Res. 51: 205. DevoTo, F. C., PERROTTO, B. M. & Estraco, V. 1970. Molar occlusal pattern of the con- temporary population of San Antonio de Los Cobres and Cobres (Argentina, Salta) and Sey and Susgues (Argentina, Jujay). J. dent. Res. 49: 997-998. DRENNAN, M. R. 1929. The dentition of a Bushman tribe. Ann. S. Afr. Mus. 24: 61-87. Drury, J. & DRENNAN, M. R. 1926. The pudendal parts of the South African Bush race. S. Afr. med. J, 22: 113-117. ERDBRINK, D. P. 1965. A quantification of the Dryopithecus and other lower molar patterns in man and some of the apes. Z. Morph. Anthrop. 57: 70-108. FriscH, J. E. 1965. Trends in the evolution of the hominoid dentition. Biblthea. primatol. 3: 1-130. GARN, S. M., DAHLBERG, A. A., Lewis, A. B. & KEREWSKyY, R. S. 1966a. Cusp number, occlusal groove pattern and human taxonomy. Nature, Lond. 210: 224-225. GARN, S. M., DAHLBERG, A. A., Lewis, A. B. & KEREWSKY, R. S. 196656. Groove pattern, cusp number and tooth size. J. dent. Res. 45: 970. Goaz, P. W. & MILLER, M. C. 1966. A preliminary description of the dental morphology of the Peruvian Indian. J. dent. Res. 45: 106-119. GOLDSTEIN, M. S. 1931. The cusps in the mandibular molar teeth of the Eskimo. Am. J. phys. Anthrop. 16: 215-235. GOLDSTEIN, M. S. 1948. Dentition of Indian crania from Texas. Am. J. phys. Anthrop. 6: 63-84. GreGcory, W. K. 1916. Studies on the evolution of the Primates. Bull. Am. Mus. nat. Hist. 35: 239-355. Grecory, W. K. 1922. The origin and evolution of the human dentition. Baltimore: Williams & Wilkins. Grecory, W. K. 1926a. Palaeontology of the human dentition. Ten structural stages in the evolution of the cheek teeth. Am. J. phys. Anthrop. 9: 401-426. GREGORY, W. K. 1926b. Some critical stages in the evolution of the human dental apparatus. J. dent. Res. 6: 71-100. Grecory, W. K. 1934. Man’s place among the anthropoids. London: Oxford University Press. GREGORY, W. K. & HELLMAN, M. 1926a. The dentition of Dryopithecus and the origin of man, Anthrop. Pap. Am. Mus. nat. Hist. 28: 1-128. GREGORY, W. K. & HELLMAN, M. 1926b. Palaeontology of the human dentition. Part I. The crown patterns of fossil and recent human molar teeth their meaning. Int. J. Orthod., Oral Surg., Radiog. 12: 1027-1037. GREGORY, W. K., HELLMAN, M. & Lewis, G. E. 1938. Fossil anthropoids of the Yale-Cam- bridge India expedition of 1935. Publ. Carneg. Instn 495: 1-27. GRINE, F. E. 1978. Occlusal morphology of the lower molars of the South African Negro and San (Bushman). J. Anat. 126: 667. GRINE, F. E. 1979. Report on two infant skeletons and an adult metatarsal excavated from Diana’s Vow. Occ. pap. natn. Mus. Rhod. (A) 4: 141-145. PERMANENT MOLARS OF NEGRO AND SAN Di GRINE, F. E. 198la. A new composite juvenile specimen of Australopithecus (Mammalia, Primates) from Member IV, Sterkfontein Formation, Transvaal. Ann. S. Afr. Mus. 84: 169-201. GRINE, F. E. 19815. Description of some juvenile hominid specimens from Swartkrans, Transvaal. Ann. S. Afr. Mus. 86: 43-71. GRINE, F. E. 1981c. A new juvenile hominid (Mammalia, Primates) from Member III, Krom- draai Formation, Transvaal, South Africa. Ann. Transv. Mus. (in press). GRINE, F. E. 1981d. Note on a new hominid specimen from Member III, Kromdraai Forma- tion, Transvaal. Ann. Transv. Mus. (in press). HANIHARA, K. 1956. Studies on the deciduous dentition of the Japanese and Japanese— American hybrids. III. Deciduous lower molars. J. anthrop. Soc. Nippon 64: 95-116. HANIHARA, K. 1961. Criteria for classification of crown characteristics of the human deciduous dentition. J. anthrop. Soc. Nippon 69: 27-45. HANIHARA, K. 1963. Crown characters of the deciduous dentition of the Japanese-American hybrids. Jn; BROTHWELL, D. R. ed. Dental anthropology: 105-124. London: Pergamon. HANIHARA, K. 1966. Mongoloid dental complex in the deciduous dentition. J. anthrop. Soc. Nippon 74: 61-72. HANIHARA, K. 1967. Racial characteristics of the dentition. J. dent. Res. 46: 923-926. HANIHARA, K. 1968a. Mongoloid dental complex in the permanent dentition. Proc. VIII Int. Congr. anthrop. ethnol. Sci. 1: 298-300. HANIHARA, K. 19685. Morphological pattern of the deciduous dentition in the Japanese- American hybrids. J. anthrop. Soc. Nippon 76: 114-121. HANIHARA, K. 1970. Mongoloid dental complex in the deciduous dentition with special reference to the Ainu. J. anthrop. Soc. Nippon 78: 3-17. HANIHARA, K. & MINAMIDATE, T. 1965. Tuberculum accessorium mediale internum in the human deciduous lower second molars. J. anthrop. Soc. Nippon 73: 9-19. HANIHARA, K., KUWASHIMA, T. & SAKAO, N. 1964. The ‘deflecting wrinkle’ on the lower molars of recent man. J. anthrop. Soc. Nippon 72: 1-8. HANIHARA, K., MASUDA, T., TANAKA, T. & TAMADA, M. 1975. Comparative studies of den- tition. Jn: WATENABE, S., KONDO, S. & MATSUNAGA, E. eds. JIBP Synthesis. Part 3 Anthropological and genetic studies of the Ainu: 256-264. Tokyo: University of Tokyo Press. Harris, E. F., TURNER, C. G. & UNDERWOOD, J. H. 1975. Dental morphology of living Yap Islanders, Micronesia. Peopling of the Pacific III. Archaeol. phys. Anthrop. Oceania 10: 218-234. HELLMAN, M. 1928. Racial characters in the human dentition. Part I. A racial distribution of the Dryopithecus pattern and its modifications in the lower molar teeth of man. Proc. Am. phil. Soc. 67: 157-174. HERSHKOVITZ, P. 1971. Basic crown patterns and cusp homologies of mammalian teeth. Jn: DAHLBERG, A. A. ed. Dental morphology and evolution: 95-150. Chicago: University of Chicago Press. Hooton, E. A. 1930. The Indians of Pecos Pueblo: a study of their skeletal remains. New Haven: Yale University Press. Hrpouicka, A. 1909. Skeletal remains from Arkansas and Louisiana. J. Acad. Nat. Sci., Philadelphia 14: 209-231. HrpuicKA, A. 1910. Anthropology of the Eskimo. Anthrop. Pap. Am. Mus. nat. Hist. 5: 1-25. Hrpb.icKa, A. 1924. New data on the teeth of early man and certain fossil European apes. Am. J. phys. Anthrop. 7: 109-132. JAcosB, T. 1967. Racial identification of Bronze Age human dentitions from Bali, Indonesia. J. dent. Res. 5 (suppl.): 903-910. JACOBSON, A. 1967. The Bantu dentition: a morphological and metrical study of the teeth, jaws and bony palate of several large groups of South African Bantu-speaking Negroids. Unpublished Ph.D. Thesis, University of the Witwatersrand, Johannesburg. JANZER, O. 1927. Die Zahne der Neu-Pommern. Ein Beitrag zur Anthropologie der Neu- Pommern, und zur Odontographie der Menschenrassen. Vjschr. Zahnheilk. 2: 1-123. JENKINS, T. 1974. Blood group A°2"t¥ population and family studies. Vox Sang. 26: 537-550. JENKINS, T. & CORFIELD, V. 1972. The red cell acid phosphatase polymorphism in southern Africa: population data and studies on the R, RA and RB phenotypes. Ann. hum. Genet. 35: 379-391. Diy ANNALS OF THE SOUTH AFRICAN MUSEUM JENKINS, T. & Nurse, G. T. 1974. The red cell 6-phosphogluconate dehydrogenase poly- morphism in certain southern African populations with first report of a new phenotype. Ann. hum. Genet. 38: 19-29. JENKINS, T. & STEINBERG, A. G. 1966. Some serum protein polymorphisms in Kalahari Bushmen and Bantu: gamma globulins, haptoglobins and transferrins. Am. J. hum. Genet. 18: 399-407. JENKINS, T. & Tosias, P. V. 1977. Nomenclature of population groups in southern Africa. Afr. Stud. 36: 49-55. JENKINS, T., BLECHER, S. R., SMITH, A. N. & ANDERSON, C. G. 1968. Some hereditary red cell traits in Kalahari Bushmen and Bantu: hemoglobins, glucose-6-phosphate dehydrogenase deficiency, and blood groups. Am. J. hum. Genet. 20: 229-309. JENKINS, T., ZOUTENDYK, A. & STEINBERG, A. G. 1970. Gammaglobulin groups (Gm and Inv) of various southern African populations. Am. J. phys. Anthrop. 32: 197-218. JORGENSEN, K. D. 1955. The Dryopithecus pattern in recent Danes and Dutchmen. J. dent. Res. 34: 195-208. JORGENSEN, K. D. 1956. The deciduous dentition: a descriptive and comparative odonto- logical study. Acta Odont. Scand. 14 (suppl. 20): 1-209. KaTICH, J. F. & TURNER, C. G. 1975. The dentition of prehistoric Hawaiians and the question of New World origins of Polynesians. Actas XLI Congr. Internacional de Americanistas, Mexico City, 1974, 1: 192-207. KIERNBERGER, A. 1955. Morphologische Untersuchungen am Gebiss und den Zahnen der Buschmann-Hottentotten Gruppe. Sber. Ost. Akad. Wiss. (ser. A) 10: 1-38. Kirk, R. L., PARKER, W. C. & BEARN, A. G. 1964. The distribution of transferrin variants D, and Dgyj in various populations. Acta genet. 14: 41-S1. KrutT, L. H. & SINGER, R. 1963. Steatopygia. Am. J. phys. Anthrop. 21: 181-187. LAVELLE, C. 1971. Mandibular molar tooth configurations in different racial groups. J. dent. Res. 50: 1353. Lawton, A. C. 1967. Bantu pottery of southern Africa. Ann. S. Afr. Mus. 49: 1-440. LEIGH, R. W. 1934. Notes on the somatology and pathology of ancient Egypt. Univ. Calif. Publ. Am. Archeol. Ethnol. 34: 1-54. LEIGH, R. W. 1937. Dental morphology and pathology in pre-Spanish Peru. Am. J. phys. Anthrop. 22: 267-296. LOMBARDI, A. V. 1975. Tooth size associations of three morphologic dental traits in a Mela- nesian population. J. dent. Res. 54: 239-243. MACINTYRE, G. T. 1966. The Miacidae (Mammalia, Carnivora). Part I. The systematics of Ictidopappus and Protictis. Bull. Am. Mus. nat. Hist. 131: 115-210. MALHERBE, M. & OckKERSE, T. 1944. Dental caries in a high and low incidence area in South Africa. A study of possible contributory factors with special reference to diet. S. Afr. J. med. Sci. 9: 75-88. MatTsuDA, T. 1961. Studies on the Dryopithecus pattern of Japanese residing in Hokuriku district. Folia Anat. Jap. 37: 317-330. Mo c ter, I. J. 1965. Dental fluorose og caries. Copenhagen: Rhodos. MOL er, I. J. 1967. Influence of microelements on the morphology of the teeth. J. dent. Res. 46: 933-937. Moorrees, C. F. A. 1950. The dentition as a criterion of race with special reference to the Aleut. J. dent. Res. 30: 815-821. Moorrees, C. F. A. 1957. The Aleut dentition: a correlative study of dental characteristics in an Eskimoid people. Cambridge: Harvard University Press. Morais, D. H. 1970. On deflecting wrinkles and the Dryopithecus pattern in human mandibular molars. Am. J. phys. Anthrop. 32: 97-104. NELSON, C. T. 1938. The teeth of the Indians of Pecos Pueblo. Am. J. phys. Anthrop. 23: 261-293. Nig, N. H., Hut, C. H., JENKINS, J. G., STEINBRENNER, K. & BENT, D. H. 1975. Statistical package for the social sciences. 2nd ed. New York: McGraw-Hill. Nurse, G. T. & JENKINS, T. 1977. Health and the hunter-gatherer: biomedical studies on the hunting and gathering populations of southern Africa. Monographs in human genetics 8: 1-126. Basel: S. Karger. OckeErRSE, T. 1943. The chemical composition of enamel and dentin in high and low caries areas in South Africa. J. dent. Res. 22: 441-446. PERMANENT MOLARS OF NEGRO AND SAN 213 OskuTu, R. O., Nurse, G. T. & JENKINS, T. 1977. Red cell enzyme polymorphism in the Yoruba. Hum. Hered. 27: 444-453. ORANJE, P. 1934. The dentition of the Bush race. S. Afr. J. Sci. 31: 576. OsBorn, H. F. 1888. The nomenclature of mammalian molar cusps. Am. Nat. 22: 926-928. PEDERSEN, P. O. 1949. The East Greenland Eskimo dentition: numerical variations and anatomy, a contribution to comparative ethnic odontography. Copenhagen: C. A. Reitzels. PERZIGIAN, A. J. 1976. The dentition of the Indian Knoll skeletal population: odontometrics and cusp number. Am. J. phys. Anthrop. 44: 113-122. REMANE, A. 1952. Der vordere Pramolar (P3) von Australopithecus prometheus und die mor- phologische Stellung des Australopithecinengebisses. Z. Morph. Anthrop. 43: 288-310. REMANE, A. 1960. Zahne und Gebiss. Jn: HOFer, A., SCHULTZ, A. H. & Starck, D. eds. Primatologia 3: 637-846. RIGHTMIRE, G. P. 1970. Bushman, Hottentot and Negro crania studied by distance and dis- crimination. Am. J. phys. Anthrop. 33: 169-196. ROBINSON, J. T. 1956. The dentition of the Australopithecinae. Mem. Transy. Mus. 9: 1-179. ROSENZWEIG, K. A. & ZILBERMAN, Y. 1967. Dental morphology of Jews from Yemen and Cochin. Am. J. phys. Anthrop. 26: 15-22. ROSENZWEIG, K. A. & ZILBERMAN, Y. 1969. Dentition of the Bedouin of Israel. II. Morphology. Am. J. phys. Anthrop. 31: 199-204. ROSENZWEIG, K. A., MAss, E. & SMITH, P. 1969. La denture des Samaritans. Bull. Grpmt int. Rech. scient. Stomat. 12: 95-106. SAKURA, H. 1979. Variations of fovea anterior in lower molars among some fossil and recent hominids. Bull. natn. sci. Mus. Tokyo (D, Anthrop.) 5: 1-11. SANTACHIARA-BENERECETTI, A. S., RANZANI, G. N. & ANTONINI, G. 1977. Studies on African pygmies. V. Red cell acid phosphatase polymorphism in Babinga pygmies — high frequency of ACPR allele. Am. J. hum. Genet. 29: 635-638. SCHULTZE-JENA, L. 1928. Zur Kenntnis des KOrpers der Hottentotten und Buschmdnner. Jenaische Denkschriften 17: 147-228. SCHWALBE, G. 1916. Uber den fossilen Affen Oreopithecus bambolii. Z. Morph. Anthrop. 19: 149-254. SELENKA, E. 1898. Rassen, Schadel und Bezahnung des Orangutan. Menschenaffen 1: 1-91. SENYUREK, M. S. 1952a. The dentition of the Chalcolithic and Copper Age inhabitants of Anatolia. Part I. A morphological study of the permanent lower molars of the Chalco- lithic and Copper Age inhabitants of Anatolia. Revue Fac. Lang., Hist., Geogr., Univ. Ankara 10: 57-77. SENYUREK, M. S. 19526. A study of the dentition of the ancient inhabitants of Alaca Héyiik. Belletin 16: 153-224. SHAw, J. C. M. 1927. Cusp development on the second lower molar of the Bantu and Bush- man. Am. J. phys. Anthrop. 11: 97-100. SINGER, R. 1978. The biology of the San. Jn: Tostas, P. V. ed. The Bushmen: San hunters and herders of southern Africa: 115-129. Cape Town: Human & Rousseau. SINGER, R. & WEINER, J. S. 1963. Biological aspects of some indigenous African populations. SWest. J. Anthrop. 19: 162-176. SKARYD, S. M. 1971. Trends in the evolution of the pongid dentition. Am. J. phys. Anthrop. 35: 223-240. SLoME, D. 1929. The osteology of a Bushman tribe. Ann. S. Afr. Mus. 24: 33-60. SNYDER, R. G., DAHLBERG, A. A., SNOW, C. C. & DAHLBERG, T. 1969. Trait analysis of the dentition of the Tarahumara Indians and Mestizos of the Sierra Madre Occidental, Mexico. Am. J. phys. Anthrop. 31: 65-76. SOFAER, J. A., MACLEAN, C. J. & BaiLit, H. L. 1972. Heredity and morphological variation in early and late developing human teeth of the same morphological class. Archs oral Biol. 17: 811-816. SOFAER, J. A., NISWANDER, J. D., MACLEAN, C. J. & WORKMAN, P. L. 1972. Population studies on southwestern Indian tribes. V. Tooth morphology as an indicator of biological distance. Am. J. phys. Anthrop. 37: 357-366. STEINBERG, A. G., JENKINS, T., NURSE, G. T. & HARPENDING, H. C. 1975. Gamma globulin groups on the Khoisan peoples of southern Africa: evidence for polymorphism for a Gm?» >, 33, 14,21 haplotype among the San. Am. J. hum. Genet. 27: 528-542. 214 ANNALS OF THE SOUTH AFRICAN MUSEUM STESLICKA, W. 1948. Wzor dryopitekoidalny na molarach Zuchwy u czlowieka. Annls Univ. Mari. Curie-Sklodowska 3: 195-222. SULLIVAN, L. R. 1920. Differences in the pattern of the second lower molar tooth. Am. J. phys. Anthrop. 3: 25. SUZUKI, M. & SAkal, T. 1956a. On the *Dryopithecus pattern’ in recent Japanese. J. anthrop. Soc. Nippon 64: 87-94. SUZUKI, M. & Sakal, T. 19565. On the ‘deflecting wrinkle’ in recent Japanese. J. anthrop. Soc. Nippon 65: 49-53. SUZUKI, M. & SAKAI, T. 1956c. On the ‘tuberculum accessorium mediale internum’ in recent Japanese. J. anthrop. Soc. Nippon 64: 95-107. SUZUKI, M. & SAKAI, T. 1957. The living Sakhalin Ainu dentition. Anthrop. Rep. 18: 303-346. SuZuUKI, M. & SAKAI, T. 1973. Occlusal surface morphology of lower molars and the second deciduous molars among the living Polynesians. Am. J. phys. Anthrop. 39: 305-316. SZALAY, F. S. 1969. Mixodectidae, Microsyopidae and the insectivore-primate transition. Bull. Am. Mus. nat. Hist. 140: 193-330. SZLACHETKO, K. 1959. Investigations on the morphology of the human deciduous dentition. Acta Fac. Rerum nat., Univ. Comen., Bratisl. 3: 247-279. TAKEHISA, S. 1957. Anatomical and anthropological studies on the teeth of American white races and Japanese. Shika Gakuho 57: 1-41. THOMSEN, S. 1955. Dental morphology and occlusion in the people of Tristan da Cunha. Oslo: Det Norske Videnkaps-Akadem1i. Tosias, P. V. 1955-6. Les Bochimans Auen et Naron de Ghanzi. Contribution a l’étude des “Anciens Jaunes’ sud-africains. I-IV. Anthropologie 59: 235-252, 429-462; 60: 22-52, 268-289. TosiAs, P. V. 1960. Bushman hunter—gatherers: a study in human ecology. In: Davies, D. H. S. ed. Ecological studies in southern Africa: 67-86. The Haag: W. Junk. TosiAs, P. V. 1961. Fingerprints and palmer prints of Kalahari Bushmen. S. Afr. J. Sci. 57: 333-345, TosIAs, P. V. 1966. The peoples of Africa south of the Sahara. Jn: BAKER, P. T. & WEINER, J. S. eds. The biology of human adaptability: 111-200. Oxford: Clarendon Press. TosiAs, P. V. 1971. Human skeletal remains from the Cave of Hearths, Makapansgat, Northern Transvaal. Am. J. phys. Anthrop. 34: 335-368. Tosias, P. V. 1972. Recent human biological studies in southern Africa, with special reference to Negroes and Khoisans. Trans. R. Soc. S. Afr. 40: 109-133. Tosias. P. V. 1978. Introduction to the Bushmen or San. Jn: Tostas, P. V. ed. The Bushmen: San hunters and herders of southern Africa: 1-15. Cape Town: Human & Rousseau. TURNER, C. G. 1970. New classifications of non-metrical dental variation. Am. J. phys. Anthrop. 33: 144-145. TURNER, C. G. 1976. Dental evidence on the origin of the Ainu and Japanese. Science 193: 911-913. TURNER, C. G. 1979. Dental anthropological indications of agriculture among the Jomon people of central Japan. X. Peopling of the Pacific. Am. J. phys. Anthrop. 51: 619-636. TURNER, C. G. & HANIHARA, K. 1977, Additional features of the Ainu dentition. V. Peopling of the Pacific. Am. J. phys. Anthrop. 46: 13-24. TURNER, C. G. & ScoTT, G. R. 1976. The dentition of living Easter Islanders, eastern Poly- nesia. Peopling of the Pacific. I. In: DAHLBERG, A. A. & GRABER, T. A. eds. Orofacial growth and development: 45-84. Chicago: Aldine. TURNER, C. G. & SWINDLER, D. R. 1978. The dentition of New Britain West Nakanai Mela- nesians. VIII. Peopling of the Pacific. Am. J. phys. Anthrop. 49: 361-372. TURNER, C. G., Scott, G. R. & LARSEN, M. 1970. Mandibular molar cusp 7 plaque and definitions of variations. Dep. Anthrop., Arizona State University, Tempe, Arizona. (Roneoed.) TURNER, C. G., Scott, G. R. & Rose, T. A. 1969. Mandibular molar cusp 6 plaque and definitions of variation. Dept. Anthrop., Arizona State University, Tempe, Arizona. (Roneoed.) VAN REENEN, J. F. 1964. Dentition, jaws and palate of the Kalahari Bushmen. J. dent. Ass. S. Afr. 19: 1-37. VAN REENEN, J. F. 1966. Dental features of a low-caries primitive population. J. dent. Res. 45: 703-713. PERMANENT MOLARS OF NEGRO AND SAN 2US5 VAN VALEN, L. 1966. Deltatheridia, a new order of mammals. Bull. Am. Mus. nat. Hist. 132: 1-126. VON KOENIGSWALD, G. H. R. 1952. Gigantopithecus blacki von Koenigswald, a giant fossil hominoid from the Pleistocene of South China. Anthrop. Pap. Am. Mus. nat. Hist. 43: 295-325. WANG, A. C., SUTTON, H. E. & Scott, I. D. 1967. Transferrin D,: identity in Australian aborigines and American Negroes. Science 156: 936-937. WEIDENREICH, F. 1937. The dentition of Sinanthropus pekinensis: a comparative odontography of the hominids. Palaeont. sin. (n.s. D, 1) 101: 1-180 & 1-120. WEIDENREICH, F. 1945. Giant early man from Java and south China. Anthrop. Pap. Am. Mus. nat Hist. 40: 1-134. WEINER, J. S., AINSWORTH HARRISON, G. SINGER, R., HARRIS, R, & Jop., W. 1964. Skin colour in southern Africa. Hum. Biol. 36: 294-307. WESTPHAL, E. O. 1963. The linguistic prehistory of southern Africa: Bush, Kwadi, Hottentot and Bantu linguistic relationships. Africa 33: 237-265. re _ ie re, hs . = y ; & 3) i i 1, F he i , Bi A ‘ \ ) 7 i f a 4 , "a - i, ae an 4: : ; , ; ie erry, 6. SYSTEMATIC papers must conform to the International code of zoological nomenclature (particularly Articles 22 and 51). Names of new taxa, combinations, synonyms, etc., when used for the first time, must be followed by the appropriate Latin (not English) abbreviation, e.g. gen. nov., sp. nov., comb. nov., Syn. nov., etc. ‘An author’s name when cited must follow the name of the taxon without intervening punctuation and not be abbreviated; if the year is added, a comma must separate author’s name and year. The author’s name ‘(and date, if cited) must be placed in parentheses if a species or subspecies is transferred from its original genus. The name of a subsequent user of a scientific name must be separated from the scientific name by a colon. Synonymy arrangement should be according to chronology of names, i.e. all published scientific names by which the species previously has been designated are listed in chronological order, with all references to that name following in chronological order, e.g.: Family Nuculanidae Nuculana (Lembulus) bicuspidata (Gould, 1845) Figs 14-15A Nucula (Leda) bicuspidata Gould, 1845: 37. Leda plicifera A. Adams, 1856: 50. Laeda bicuspidata Hanley, 1859: 118, pl. 228 (fig. 73). Sowerby, 1871: pl. 2 (fig. 8a—b). Nucula largillierti Philippi, 1861: 87. Leda bicuspidata: Nicklés, 1950: 163, fig. 301; 1955: 110. Barnard, 1964: 234, figs 8-9. Note punctuation in the above example: comma separates author’s name and year semicolon separates more than one reference by the same author full stop separates references by different authors figures of plates are enclosed in parentheses to distinguish them from text-figures dash, not comma, separates consecutive numbers Synonymy arrangement according to chronology of bibliographic references, whereby the year is placed in front of each entry, and the synonym repeated in full for each entry, is ~ not acceptable. In describing new species, one specimen must be designated as the holotype; other speci- mens mentioned in the original description are to be designated paratypes; additional material not regarded as paratypes should be listed separately. The complete data (registration number, depository, description of specimen, locality, collector, date) of the holotype and paratypes must be recorded, e.g.: Holotype SAM-—A13535 in the South African Museum, Cape Town. Adult female from mid-tide region, King’s Beach Port Elizabeth (33°51’S 25°39’E), collected by A. Smith, 15 January 1973. Note standard form of writing South African Museum registration numbers and date. 7. SPECIAL HOUSE RULES Capital initial letters (a) The Figures, Maps and Tables of the paper when referred to in the text b] e.g. ‘... the Figure depicting C. namacolus ...’; *. . . in C. namacolus (Fig. 10)...’ (b) The prefixes of prefixed surnames in all languages, when used in the text, if not preceded by initials or full names e.g. Du Toit but A.L.du Toit; Von Huene but F. von Huene (c) Scientific names, but not their vernacular derivatives e.g. Therocephalia, but therocephalian Punctuation should be loose, omitting all not strictly necessary Reference to the author should be expressed in the third person Roman numerals should be converted to arabic, except when forming part of the title of a book or article, such as “Revision of the Crustacea. Part VIII. The Amphipoda.’ Specific name must not stand alone, but be preceded by the generic name or its abbreviation -to initial capital letter, provided the same generic name is used consecutively. Name of new genus or species is not to be included in the title: it should be included in the abstract, counter to Recommendation 23 of the Code, to meet the requirements of Biological Abstracts. F. E. GREENE OCCLUSAL MORPHOLOGY OF THE MANDIBULAR PERMANENT MOLARS OF THE SOUTH AFRICAN NEGRO AND THE KALAHARI SAN (BUSHMAN) OH ie | $67X \RT 6 DECEMBER 1981 _ ISSN 0303-2515 oe | FEB 171982 ~~ _LIBRARIES a ~~ MUSEUM — CAPE TOWN. INSTRUCTIONS TO AUTHORS 1. MATERIAL should be original and not published elsewhere, in whole or in part. 2. LAYOUT should be as follows: (a) Centred masthead to consist of Title: informative but concise, without abbreviations and not including the names of new genera or species Author’s(s’) name(s) Address(es) of author(s) (institution where work was carried out) Number of illustrations (figures, enumerated maps and tables, in this order) (b) Abstract of not more than 200 words, intelligible to the reader without reference to the text (c) Table of contents giving hierarchy of headings and subheadings (d) Introduction (e) Subject-matter of the paper, divided into sections to correspond with those given in table of contents (f) Summary, if paper is lengthy (g) Acknowledgements (h) References (i) Abbreviations, where these are numerous 3. MANUSCRIPT, to be submitted in triplicate, should be typewritten and neat, double spaced with 2,5 cm margins all round. First lines of paragraphs should be indented. Tables and a list of legends for illustrations should be typed separately, their positions indicated in the text. All pages should be numbered consecutively. Major headings of the paper are centred capitals; first subheadings are shouldered small capitals; second subheadings are shouldered italics; third subheadings are indented, shouldered italics. Further subdivisions should be avoided, as also enumeration (never roman numerals) of headings and abbreviations. Footnotes should be avoided unless they are short and essential. Only generic and specific names should be underlined to indicate italics; all other marking up should be left to editor and publisher. 4. ILLUSTRATIONS should be reducible to a size not exceeding 12 « 18 cm (19 cm including legend); the reduction or enlargement required should be indicated; originals larger than °35 x 47 cm should not be submitted; photographs should be rectangular in shape and final size. A metric scale should appear with all illustrations, otherwise magnification or reduction should be given in the legend; if the latter, then the final reduction or enlargement should be taken into consideration. All illustrations, whether line drawings or photographs, should be termed figures (plates are not printed; half-tones will appear in their proper place in the text) and numbered in a single series. Items of composite figures should be designated by capital letters; lettering of figures is not set in type and should be in lower-case letters. The number of the figure should be lightly marked in pencil on the back of each illustration. 5. REFERENCES cited in text and synonymies should all be included in the list at the end of the paper, using the Harvard System (ibid., idem, loc. cit., op. cit. are not acceptable): (a) Author’s name and year of publication given in text, e.g.: ‘Smith (1969) describes .. .’ ‘Smith (1969: 36, fig. 16) describes... .’ ‘As described (Smith 1969a, 19695; Jones 1971)’ ‘As described (Haughton & Broom 1927)...’ ‘As described (Haughton et al. 1927)...’ Note: no comma separating name and year Dagination indicated by colon, not p. names of joint authors connected by ampersand et al. in text for more than two joint authors, but names of all authors given in list of references. (b) Full references at the end of the paper, arranged alphabetically by names, chronologically within each name, with suffixes a, b, etc. to the year for more than one paper by the same author in that year, e.g. Smith (1969a, 19695) and not Smith (1969, 1969a). For books give title in italics, edition, volume number, place of publication, publisher. For journal article give title of article, title of journal in italics (abbreviated according to the World list o, scientific periodicals. 4th ed. London: Butterworths, 1963), series in parentheses, volume number, part number (only if independently paged) in parentheses, pagination (first and last pages of article). Examples (note capitalization and punctuation) BULLOUGH, W. S. 1960. Practical invertebrate anatomy. 2nd ed. London: Macmillan. FISCHER, P.—H. 1948. Données sur la résistance et de le vitalité des mollusques. J. Conch., Paris 88: 100-140. FIscHER, P.-H., DuvAL, M. & Rarry, A. 1933. Etudes sur les échanges respiratoires des littorines. Archs Zool. exp. gén. 74: 627-634. \ Konn, A. J. 1960a. Ecological notes on Conus (Mollusca: Gastropoda) in the Trincomalee region of Ceylon. Ann. Mag. nat. Hist. (13) 2: 309-320. Konn, A. J. 19606. Spawning behaviour, egg masses and larval development in Conus from the Indian Ocean. Bull. Bingham oceanogr. Coll. 17 (4): 1-51. THIELE, J. 1910. Mollusca: B. Polyplacophora, Gastropoda marina, Bivalvia. In: SCHULTZE, L. Zoologische und anthropologische Ergebnisse einer Forschungsreise im westlichen und zentralen Siid-Afrika 4: 269-270. Jena: Fischer. Denkschr. med.-naturw. Ges. Jena 16: 269-270. (continued inside back cover) ANNALS OF THE SOUTH AFRICAN MUSEUM ANNALE VAN DIE SUID-AFRIKAANSE MUSEUM Volume 86 #Band December 1981 Desember Part 6 Deel A HYENA-ACCUMULATED BONE ASSEMBLAGE FROM LATE HOLOCENE DEPOSITS AT DEELPAN, ORANGE FREE STATE By L2 SCOT & Re-G KEBIN Cape Town Kaapstad The ANNALS OF THE SOUTH AFRICAN MUSEUM are issued in parts at irregular intervals as material becomes available Obtainable from the South African Museum, P.O. Box 61, Cape Town 8000 Die ANNALE VAN DIE SUID-AFRIKAANSE MUSEUM word uitgegee in dele op ongereelde tye na gelang van die beskikbaarheid van stof Verkrygbaar van die Suid-Afrikaanse Museum, Posbus 61, Kaapstad 8000 OUT OF PRINT/UIT DRUK IL GES, S83), HOLD, 4-8, & tani), SS, S, EO, GA, to), CD), B TOD, , HOES, GED, SW, ejay), ICS), WO), M7, BES), 32G), 33, ZO EDITOR/REDAKTRISE Ione Rudner Copyright enquiries to the South African Museum Kopieregnavrae aan die Suid-Afrikaanse Museum ISBN 0 86813 027 3 Printed in South Africa by In Suid-Afrika gedruk deur The Rustica Press, Pty., Ltd., Die Rustica-pers, Edms., Bpk., Court Road, Wynberg, Cape Courtweg, Wynberg, Kaap A HYENA-ACCUMULATED BONE ASSEMBLAGE FROM LATE HOLOCENE DEPOSITS AT DEELPAN, ORANGE FREE STATE By Louis SCOTT Institute for Environmental Sciences, University of the Orange Free State & RICHARD G. KLEIN Department of Anthropology, University of Chicago (With 3 figures and 2 tables) [MS accepted 5 August 1981] ABSTRACT Late Holocene sands fringing Deelpan, Western Orange Free State, have yielded a sample of bones derived from steenbok, springbok, black wildebeest, blesbok, quagga or Burchell’s zebra, caracal or serval, slender mongoose, clawless otter, black-backed jackal, and eagle or hawk, as well as coprolites of a hyena. The composition of the fauna is in keeping with pollen evidence that highveld vegetation prevailed at the time the bones accumulated. The absence of artefacts and porcupine gnaw marks, in combination with the hyena coprolites and two bones almost certainly damaged by hyena teeth, indicate that hyenas were responsible for the bone accumulation. Like other hyena-accumulated samples, the Deelpan one differs from human (archaeological) samples in the relatively high number of carnivore individuals represented and in the tendency for the ratio of postcranial to cranial bones to increase with the size of the species involved. CONTENTS PAGE INtROGUCHIOMare es ces Verte ose ee cia ae a ae AG, SE Uti tN ot ae ch oem ee UN ee a MORI Ries eet ee tages 218 Contextof the: boneassemblageh sa) 2554-45 555s eae 219 sine bonevassemblagec iia as he ae eek Sessu sere 220 AXCKnOWledSements) 24. eines ee eee ee ee 226 INCLETENM CES Heke sss ea se esi acne eos cakes, SP Ae ee 226 INTRODUCTION In the course of a multidisciplinary investigation into the history and origin of pans in the Orange Free State, Scott recovered an assemblage of bones from unconsolidated sandy deposits on the eastern margin of Deelpan (Honing Kopje Pan) (approximately 29°11’S 25°45'E), between Bloemfontein and Petrusburg (Fig. 1). Artefacts were absent, and nine associated coprolites, together with damage marks on two of the bones, point to hyenas as the bone collectors. In this brief report, Scott describes the setting and sedimentary 217 Ann. S. Afr. Mus. 86 (6), 1981: 217-227, 3 figs, 2 tables. 218 ANNALS OF THE SOUTH AFRICAN MUSEUM to Petrusburg o Bloemfontein B Fig. 1. Left: The location of the Deelpan fossil site within South Africa. Right: A schematic plan of Deelpan, showing the distribution of the sandy deposits on its southern and eastern margins and the points (Deelpan A and B) where bones were recovered. context of the bone assemblage, while Klein discusses its implications for distinguishing hyena bone accumulations from human (archaeological) ones. In a future paper, K. W. Butzer (University of Chicago) will analyse a suite of sediment samples from the site, with the goal of elucidating its climatic and environmental history. SETTING Deelpan is typical of the numerous pans or playas that dot the plains of the western Orange Free State (De Bruiyn 1971, 1972; Le Roux 1978). The pans are thought to result from the interaction of bedrock geology, slope, salt concentration, chemical weathering, seasonal climate, and animal activity, the combined effect of which is to foster deflation. The prevailing westerly winds then build sand-dunes along the south-eastern and eastern margins of most pans in the region (De Bruiyn 1971, 1972). The open plains of the Orange Free State are underlain by typical Karoo sedimentary rocks intruded by dolerites that form rocky hills (‘koppies’). Annual precipitation in the Deelpan area is less than 500 mm. The vegetation belongs to the so-called False Upper Karoo (Acocks 1953, veld type 36), comprising grassland with a high proportion of Karoo shrubs, especially Com- positae. Woody species such as Olea africana, Rhus lancea, and Rhus erosa occur on dolerite ‘koppies’. The pan floor of approximately 6,5 km? does not support any vegetation and consists of clays rich in carbonates and other salts HYENA-ACCUMULATED BONE ASSEMBLAGE 219 concentrated on the surface through evaporation. During most of the year, the pan is dry, except for the south-western portion where perennial springs occur. However, during especially rainy years, water may fill the pan to a depth of 2 m and remain for a year or more. When filled, the pan attracts countless numbers of water-birds. In prehistoric times, it probably also attracted large numbers of antelopes and other animals. Artefacts found along the margins of Deelpan and various Other pans in the region point to frequent visits by prehistoric man (Butzer et al. 1973; Butzer 1974; Horowitz et al. 1978). CONTEXT OF THE BONE ASSEMBLAGE As at many western Free State pans, unconsolidated Quaternary deposits at Deelpan occur along its southern and eastern margins. Recent erosion of these deposits has allowed recognition of a three-part sedimentary sequence (from top to bottom): 1, more than 2 m of semi-compacted sands, exhibiting distinct stratification in places; 2, approximately 1 m of semi-compacted, non- stratified sands, especially obvious in the southernmost exposures; and 3, more than 3 m of compact, calcareous clayey sands. The bones described here came from near the bottom of the uppermost unit (1). Only traces of bones were found in the two lower units, along with occasional mollusc shells and frag- ments of ostrich egg-shell. Within the uppermost sedimentary unit, bones were found at the same level in two distinct concentrations, labelled A and B, approximately 3 m apart. At point B, a subtle colour difference suggests the bones lie in a pocket roughly 30 cm across within the loose sands. It seems likely that the pocket is the cross-section of an ancient burrow, quite possibly part of a system also including the bone occurrence at A. This hypothesis can be checked only by excavation, which may also lead to the recovery of bones in the sands between A and B. The present bone collection consists totally of pieces that were eroding out of the sections at A and B or had already eroded out and were lying on the pediment below. Preliminary observations indicate that the sedimentary sequence observed at Deelpan also characterizes the unconsolidated sediments found at most isolated pans in the western Orange Free State. The similarity almost certainly reflects a common response to past climatic change. Horowitz et al. (1978) have described an especially similar sequence at Voigtspost Pan, approximately 40 km west of Deelpan. The specific palaeoclimatic events involved at Voigts- post remain to be worked out, but comparison to Deelpan is still informative. Particularly pertinent are *C determinations of 6350 +75 B.P. (Pta — 1520) and 1220 + 50B.P. (Pta — 1483) on ostrich egg-shell fragments from sediments at Voigtspost equivalent to Deelpan sedimentary units 3 and 1 respectively. The dates indicate that the visible sedimentary sequences at Voigtspost and Deelpan date from the Holocene, while the bone assemblage from Deelpan A and B is probably of late Holocene age. 220 ANNALS OF THE SOUTH AFRICAN MUSEUM Limited palynological evidence from Voigtspost suggests that the middle unit (equivalent to unit 2 at Deelpan) was deposited under slightly moister conditions than the underlying and overlying sediments (equivalent to units 3 and 1 at Deelpan). However, the pollen spectra imply basically highveld vegetation throughout the sequence. The pollen from the uppermost sands at Voigtspost, equivalent to the bone-bearing unit at Deelpan, comprises 2% Gramineae, 43% Chenopodiaceae, 11% Compositae, 30% spores of Riccia and Ophioglossum, and traces of other plants. Pollens were not found in the equivalent Deelpan sediments, but 90% of the pollen found in coprolites accompanying the Deelpan bones derived from Gramineae. The difference probably reflects the differing nature of sediments and coprolites as pollen traps, plus the fact that the pollen in the Voigtspost samples was probably deposited over a relatively long period in the halophytic environment of the pan shore. Like the Voigtspost pollen, that in the Deelpan coprolites almost certainly reflects essentially highveld vegetation. THE BONE ASSEMBLAGE Since there is every likelihood that the bones from Deelpan A and B were accumulated at the same time by the same agency, they have been treated here as a single assemblage. The bones are relatively fresh looking, with no apparent mineralization, in keeping with their probable late Holocene age. Fragmenta- tion 1s minimal and nearly every piece was immediately identifiable to skeletal part and species. The species represented are listed in Table 1, which also presents the number of bones assigned to each and the minimum number of individuals from which the bones derive. The skeletal parts by which each species is represented are listed in Table 2. The ungulate species present are the classic, historic inhabitants of the highveld, and their occurrence is thus totally in keeping with pollen evidence that highveld vegetation prevailed at the time the TABLE 1 The number of identifiable bones: the minimum number of individuals by which various species are represented in the bone assemblage from Deelpan A and B. Steenbok (Raphicerus campesins) @ 1 sere eee eee Sane nee 1/1 Springbok) (Anidorcas marsupials) meaner eee ae: oe eae eee 48/3 black wildebeest (Connochacites enow)r) ener ee ee ee eee 24/4 blesboks(Damaliscus doncaster os te: ine pene iar eo eae ne aie v2 quagga or Burchell’s zebra (Equus quagga or E. burchelli)................. 1/1 Caracal or serval (Kelisicanacal Or asenval) sn en eee eee 3/1 hyena (Hyacnidac gensetsp-wimdet.) =e one eee ee a no ee 9 coprolites slender mongoose (Henpestesisanguineus) 2. 458s 4 eee ae 1/1 clawless otter:(Alomy%Capensts) ans cis.ch enya cao Ne wioe aye eG 2/1 black-backed jackalli(Ganisimesomelas) a. sone is a ie ae 6/2 eagle or hawk (Accipitridae gen. et sp. indet.) ................-0..- 00 eee 2/1 Mepel| HYENA-ACCUMULATED BONE ASSEMBLAGE V/I I/II W/E JOGO 00000000N sosuvyjeyd L/T I/I L/I Doouogodn DOO OOO sjesieye}oul /Z O.8 0.0 bao oO sjesie} Joyo V/I bo UDO oOO DON OW winoursjeo VI VI SooubovuDponoo ad snyjeserse coodd00D DOU DDD Omo eyjaved V/I Z/¢ 600060000 60 0.010000 60 6.0 viqn L/I aapis) felkoleey (eo otiloccutentenelerteltelice Bultic)! V/I L/I LI Oo Ooo OO lsu Gsbr0 9} euUTWIOUUT L/I ZT p/p Z/S Glo ot 0.0 GO ODD sjedieorjou VI L/I GOaggonoonodbo0 000 sjedieo L/I Wik B1d10 0-0 040-0) 0 0100.0. Cd. 2080.0 euyn 2/7 Z/¢ Z/P 5160 010" 01010 0.00 0.00070 snipes 1/1 Wt mn E/E Zig Se ea snroumny L/I Z/¢ Goancoooon ad ond 000 vindeos Wiha on eed Se squ WiG¥st kee et ee ee IPIQIJIOA [BpNned ooo 00.0 OVIGIIJIOA [BIOVS Aeceaencateed OBIGIIOA Jequin] ma! 1/2 Wits 2 2 i See IVIQIIOA D19eIOY} W ip) ee eee sjeorarao 10410 Z/Z hat ehnencscunova: sixe UI i ee Se cert sepe 7/7 L/Z L/I Zip he a aetna d]qIpuew UZ ae cee eyxeus VI ee Febonbasabeann: gjewi9y tie gjeul Se es oS jopuoy pie jeyort iano — esoo8uom ~—- FPA es as yoqss{q ysooqepyim yoqsutds yoquadys /ayseo jyeoeres =—seddenb ‘gq pue v uedjs0q Wor euney oy) UI UOXe) Jod yUOWIIA JeJO;Oys Jod s[eNprIAIpU! Jo JoquINU WiNUITUTU 9Y}/souOg s[QeyNHUspt Jo Joquinu oy (ACOCKAD 222 ANNALS OF THE SOUTH AFRICAN MUSEUM bones accumulated. The species composition is also totally consistent with a late Holocene age. The complete absence of domestic stock (cattle and/or sheep and goats) further suggests that the bones antedate European settlement of the area, beginning in the middle part of the last century. It is possible that domestic stock were introduced to the area even earlier by indigenous Khoi 1 oR ae ny: iD S ag A “L a if fin ? Si Sty sa YS “esr: Eee (f BY ye + a MAY, iy f De 2 a 3 Sy 3A, BS] th y: aU 7 2 a bd) ~ - RS re ¢ ef 2. Damaliscus dorcas Hyaenid coprolites (12,39; 8,4g) ZT a ae CM Fig. 2. Left: A blesbok ulna from Deelpan A displaying punctate depressions probably caused by the canines of a hyena. Right: Hyena coprolites from Deelpan A. HYENA-ACCUMULATED BONE ASSEMBLAGE 223 Connochaetes gnou DEELPAN Fig. 3. A black wildebeest distal metacarpal from Deelpan A exhibiting damage to the outer table that was probably caused by hyena chewing. 224 ANNALS OF THE SOUTH AFRICAN MUSEUM herders or Iron Age mixed farmers, but Deelpan lies in a relatively dry region where both archaeological and historic records suggest that prehistoric occupa- tion was largely, if not exclusively limited to hunter—-gatherers (Maggs 1976). In southern Africa, there are three potential accumulators of bones at a site such as Deelpan. Two—people and porcupines (Hystrix africaeaustralis )}— are ruled out by the absence of their telltale signs—artefacts and gnaw marks respectively. The third—hyenas—are clearly indicated by the presence of their coprolites (9 in total) and of two ungulate bones with what are almost certainly damage marks from carnivore teeth (Figs 2-3). Both the brown hyena (Hyaena brunnea) and the spotted hyena (Crocuta crocuta) occurred in the region of Deelpan historically, and both are known to accumulate bones in dens or lairs, including abandoned aardvark burrows or Other subterranean cavities of sufficient size. It is such burrows, now filled in or collapsed, that may well be represented at Deelpan. The brown form is the more prodigious bone collector of the two, largely because it brings food back to its young, while the spotted form does not. The two species also differ in the kinds of animals that they most frequently eat. The brown hyena feeds much more often on species springbok-size and smaller, reflecting its general incom- petence as a hunter and its inability to retain the carcasses of larger animals when other large predators—scavengers are also present. (See Sutcliffe (1970), Kruuk (1972), Bearder (1977), Mills (1978a), Mills & Mills (1977), and Hen- schel et al. (1979) for information on the spotted hyena; Mills (1973, 1978D), Mills & Mills (1977), Skinner (1976), and Owens & Owens (1978, 1979) for data on the brown hyena.) Tables 1 and 2 show that springbok is the most common species at Deelpan in terms of the number of identifiable bones, while black wildebeest is most common in terms of the minimum number of individuals represented. In combination with the fact that the relative abundance of these species in the ancient Deelpan environment remains unknown and perhaps unknowable, this means that the species frequencies in the present sample do not help to establish which hyena accumulated the bones. Perhaps ultimately this question will be resolved, if it becomes possible to identify the coprolites to species. Whichever hyena is responsible, the Deelpan assemblage contrasts with hominid (archaeological) bone assemblages in two important respects: 1. The relative abundance of carnivores. In minimum individual terms, carnivores constitute 31% of the animals in the Deelpan sample, while there is no archaeological sample in Africa in which the comparable figure has been found to exceed 13%. The contrast probably reflects the fact that hyenas (especially the brown species) interact with other carnivores (especially jackals) much more frequently than hunter-gatherers do. 2. At Deelpan, there is a clear tendency for larger species to be relatively better represented by postcranial bones than by cranial ones. In archaeological assemblages, there is no such trend, and the cranial:postcranial ratio does not seem to be related to the size of the species concerned. The contrast probably HYENA-ACCUMULATED BONE ASSEMBLAGE 225 reflects the fact that hyenas find it far more difficult to transport the skulls of large animals than people do, while both kinds of bone accumulators find postcranial bones about equally portable. Klein has found that the same features characterize fossil bone assem- blages collected by hyenas at other sites in southern Africa, particularly Swartklip 1 (Hendey & Hendey 1968; Klein 1975), Equus Cave (Beaumont & Shackley pers. comm.; Klein in preparation), and the Elandsfontein ‘Bone Circle’ (Inskeep & Hendey 1966; Klein unpub.). These other samples are all much larger than the Deelpan one, and the fact that the features are still clear at Deelpan is thus a measure of the strength of the pattern they represent. One feature that might be expected to characterize hyena accumulations would be a large number of bones obviously damaged by the animals’ teeth. In fact, such bones are relatively rare not only at Deelpan but also in the other fossil assemblages mentioned above and apparently in assemblages from dens where living hyenas were observed, though precise figures on damaged bones from such dens are generally lacking. (The bones found at active dens are discussed in some of the papers on hyena feeding and foraging cited above.) The implication of this is that, where there is doubt as to the human or hyenid origin of a particular bone assemblage, principal reliance will probably have to be placed on relatively subtle features such as the abundance of carnivores or the nature of the relationship between species size and cranial:postcranial bone representation. In southern Africa, the most important bone assemblages whose origins remain problematic are certainly those from the various australopithecine caves in the Transvaal. The question is basically whether the bones were brought to the caves by the australopithecines, as argued particularly by Dart (1957a, 1957b) or perhaps by a carnivore, as discussed particularly by Brain (1980, 1981). A priori, in considering the possibility that hyenas were involved, it might seem most reasonable to attack the problem by analysing bones from recent hyena dens. However, the samples from such dens tend to be very small, while the dens themselves are mostly in marginal or degraded environments. Often, many of the bones present come from domesticated species. Further- more, especially in the case of the brown hyena, which has become compara- tively rare, it is unlikely that there are very many dens left to be sampled. All this means that the study of the fossil collections is essential both to establish a convincing pattern of differences between hyena and hominid assemblages and to maintain reasonable control over environmental and ecological variables in exploring the origins of assemblages such as those from the australopithecine caves. Klein (1975) has already argued that there are marked similarities between the bone assemblage from Swartklip (and by extension from Deelpan and other hyena-accumulated samples) on the one hand and the bone assemblage from Makapansgat on the other, as partially described by Dart (1957a, 1957b) and Wells & Cooke (1956). Some fresh data on the Makapansgat carnivores 226 ANNALS OF THE SOUTH AFRICAN MUSEUM presented by Collings et al. (1975) suggest, in fact, that the bone accumulation at Makapansgat, at least from the so-called ‘Grey Breccia’ (Member 3 of Partridge 1979), probably owes its origin largely, if not entirely, to the ancestral striped hyena, Hyaena hyaena makapani. Ultimately, it should prove possible to reach reasonable closure on this issue through the analysis of more assem- blages such as the one from Deelpan and through more detailed study of the Makapansgat collection. ACKNOWLEDGEMENTS We thank G. R. de Villiers for permission to visit Deelpan, R. Verhoeven for his help in the fieldwork, K. W. Butzer for his advice on the study of the Orange Free State pans, G. Avery for identifying the bird bones from Deelpan, K. Allwarden for drawing Deelpan bones and coprolites, the National Science Foundation for financial support to Klein, and the South African Museum for providing facilities for the study and permanent storage of the Deelpan faunal collection. REFERENCES Acocks, J. P. H. 1953. Veld types of South Africa. Mem. Bot. Surv. S. Afr. 40: 1-128. BEARDER, S. K. 1977. Feeding habits of spotted hyaenas in a woodland habitat. E. Afr. Wildl. J. 15: 263-280. BRAIN, C. K. 1980. Some criteria for the recognition of bone-collecting agencies in African caves. In: Behrensmeyer, A. K. & Hill, A. P. eds. Fossils in the making: 107-130. Chicago: University of Chicago Press. BRAIN, C. K. 1981. The hunters or the hunted: an introduction to African cave taphonomy. Chicago: University of Chicago Press. Bruyn, H. DE 1971. ’n Geologiese studie van die panne in die Westelike Oranje-Vrystaat. Unpublished M.Sc. thesis, University of the Orange Free State. Brutyn, H. De 1972. Pans in the western Orange Free State. Ann. geol. Surv. 9: 121-124. Butzer, K. W. 1974. Geo-archaeological interpretation of Acheulian calc-pan sites at Doorn- laagte and Rooidam (Kimberley, South Africa). J. archaeol. Sci. 1: 1-25. ButTzer, K. W., Fock, G. J., STUCKENRATH, R. & ZiLcH, A. 1973. Palaeohydrology of Late Pleistocene Lake Alexandersfontein, Kimberley, South Africa. Nature, Lond. 243: 328-330. CoLLinGs, G. E., CRUIKSHANK, A. R. I., MaGuire, J. M. & RANDALL, R. M. 1975. Recent faunal studies at Makapansgat Limeworks, Transvaal, South Africa. Ann. S. Afr. Mus. 71: 153-165. Dart, R. A. 1957a. The Makapansgat australopithecine osteodontokeratic culture. Jn: Clark, J. D. ed. Proceedings of the Third Pan-African Congress of Prehistory (Livingstone, 1955): 161-171. London: Chatto & Windus. Dart, R. A. 1957b. The osteodontokeratic culture of Australopithecus prometheus. Transv. Mus. Mem. 10: 1-105. HENDEY, Q. B. & HENDEy, H. 1968. New Quaternary fossil sites near Swartklip, Cape Province. Ann. S. Afr. Mus. 52: 43-73. HENSCHEL, J. R., TILSON, B. & VON BLotINitTz, F. 1979. Implications of a spotted hyaena bone assemblage in the Namib Desert. S. Afr. archaeol. Bull. 24: 127-131. Horowitz, A., SAMPSON, C. G., Scott, L. & VOGEL, J. C. 1978. Analysis of the Voigtspost Site, O.F.S. S. Afr. archaeol. Bull. 33: 152-159. INSKEEP, R. R. & HENDEY, Q. B. 1966. An interesting association of bones from the Elandsfontein fossil site. In: Actes du Ve Congres Panafricain de Prehistoire et de l Etude du Quaternaire: 109-124. Tenerife: Museo Arqueologico. HYENA-ACCUMULATED BONE ASSEMBLAGE DF, KLEIN, R. G. 1975. Paleoanthropological implications of the non-archeological bone assem- blage from Swartklip 1, southwestern Cape Province, South Africa. Quaternary Res. 5: 275-288. Kruuk, H. 1972. The spotted hyena. Chicago: University of Chicago Press. LE Roux, J. S. 1978. The origin and distribution of pans in the Orange Free State. §. Afr. Geograaf. 6: 167-176. Macs, T. M. O’C. 1976. Iron Age communities of the southern highveld. Occ. Publ. Natal. Mus. 2: 1-326. Mitts, M. G. L. 1973. The brown hyaena. Afr. Wildl. 27: 150-153. Mitts, M. G. L. 1978a. The comparative socio-ecology of the Hyaenidae. Carnivore 1: 1-6. Mitts, M. G. L. 19785. Foraging behaviour of the brown hyaena (Hyaena brunnea, Thunberg, 1820) in the southern Kalahari. Z. Tierpsychol. 48: 113-141. Mitts, M. G. L. & Mitts, M. E. J. 1977. An analysis of bones collected at hyaena breeding dens in the Gemsbok National Parks (Mammalia: Carnivora). Ann. Transv. Mus. 30: 145-155. Owens, M. J. & Owens, D. D. 1978. Feeding ecology and its influence on social organization in brown hyaenas (Hyaena brunnea, Thunberg) of the Central Kalahari Desert. E. Afr. Wildl. J. 16: 113-135. Owens, M. J. & Owens, D. D. 1979. The secret society of the brown hyaenas. Afr. Wildl. 33: 26-29. PARTRIDGE, T. C. 1979. Re-appraisal of lithostratigraphy of Makapansgat Limeworks hominid site. Nature, Lond. 279: 484-488. SKINNER, J. D. 1976. Ecology of the brown hyaena Hyaena brunnea in the Transvaal with a distribution map for southern Africa. S. Afr. J. Sci. 72: 262-269. SuTcLiFFE, A. J. 1970. Spotted hyaena: crusher, gnawer, digester, and collector of bones. Nature, Lond. 227: 1110-1113. WELLs, L. H. & Cooke, H. B. S. 1956. Fossil Bovidae from the Limeworks Quarry, Makapansgat, Potgietersrus. Palaeont. afr. 4: 1-55. i” pee a ee. a a i a |) ae 7 7 ™ ame gee Bsc < nan 7 Pee. 1¥ ii Sioa ee Es a “3 ; } tai ; 7 ek = re a ; ; { ; a a 2s 7 : = 5 . iad , Leo : . ; ~ 4 - 7 a *. e E 4 m ' _—: vA i ir > a - 1 ah ' oe : ot ea Pa ee? © bs 0 Peay i oe eee | az 1 s a - : ¢ ¥ a > oa cay ney 7 S Se a is iz ; n A hes ‘ 7 th - % on : . ; f b { on cate ane 5 oe Da i i i at 1 ’ = , mi i : _ i zt ne . , , ’ } a | ~ u : i \ x‘ rs ie : : t py ( 7 ‘ ; é ; i an as Sa F i ps ; i = “i ‘ ie ; 1 . 4 ( } - 1 f ~ ‘4 I, ; Ve , ~ a * a) = 2 ( : y 9 . , \ . = . : -% ren Y = ei t 6. SYSTEMATIC papers must conform to the International code of zoological nomenclature (particularly Articles 22 and S51). Names of new taxa, combinations, synonyms, etc., when used for the first time, must be followed by the appropriate Latin (not English) abbreviation, e.g. gen. nov., sp. nov., comb. nov., Syn. nov., etc. ‘An author’s name when cited must follow the name of the taxon without intervening punctuation and not be abbreviated; if the year is added, a comma must separate author’s name and year. The author’s name ‘(and date, if cited) must be placed in parentheses if a species or subspecies is transferred from its original genus. The name of a subsequent user of a scientific name must be separated from the scientific name by a colon. Synonymy arrangement should be according to chronology of names, i.e. all published scientific names by which the species previously has been designated are listed in chronological order, with all references to that name following in chronological order, e.g.: Family Nuculanidae Nuculana (Lembulus) bicuspidata (Gould, 1845) Figs 14-15A Nucula (Leda) bicuspidata Gould, 1845: 37. Leda plicifera A. Adams, 1856: 50. Laeda bicuspidata Hanley, 1859: 118, pl. 228 (fig. 73). Sowerby, 1871: pl. 2 (fig. 8a—b). Nucula largillierti Philippi, 1861: 87. Leda bicuspidata: Nicklés, 1950: 163, fig. 301; 1955: 110. Barnard, 1964: 234, figs 8-9. Wore punctuation in the above example: comma separates author’s name and year semicolon separates more than one reference by the same author full stop separates references by different authors figures of plates are enclosed in parentheses to distinguish them from text-figures dash, not comma, separates consecutive numbers Synonymy arrangement according to chronology of bibliographic references, whereby the year is placed in front of each entry, and the synonym repeated in full for each entry, is ~ not acceptable. In describing new species, one specimen must be designated as the holotype; other speci- mens mentioned in the original description are to be designated paratypes; additional material not regarded as paratypes should be listed separately. The complete data (registration number, depository, description of specimen, locality, collector, date) of the holotype and paratypes must be recorded, e.g.: Holotype SAM-—A13535 in the South African Museum, Cape Town. Adult female from mid-tide region, King’s Beach Port Elizabeth (33°51’S 25°39’E), collected by A. Smith, 15 January 1973. Note standard form of writing South African Museum registration numbers and date. 7. SPECIAL HOUSE RULES Capital initial letters (a) The Figures, Maps and Tables of the paper when referred to in the text e.g. *... the Figure depicting C. namacolus ...’; *. . . in C. namacolus (Fig. 10)... (b) The prefixes of prefixed surnames in all languages, when used in the text, if not preceded by initials or full names e.g. Du Toit but A.L.du Toit; Von Huene but F. von Huene (c) Scientific names, but not their vernacular derivatives e.g. Therocephalia, but therocephalian Punctuation should be loose, omitting all not strictly necessary Reference to the author should be expressed in the third person Roman numerals should be converted to arabic, except when forming part of the title of a book or article, such as ‘Revision of the Crustacea. Part VIII. The Amphipoda.’ Specific name must not stand alone, but be preceded by the generic name or its abbreviation to initial capital letter, provided the same generic name is used consecutively. Name of new genus or species is not to be included in the title: it should be included in the abstract, counter to Recommendation 23 of the Code, to meet the requirements of Biological Abstracts. L. SCOTT & R. G. KLEIN A HYENA-ACCUMULATED BONE ASSEMBLAGE FROM LATE HOLOCENE DEPOSITS AT DEELPAN, ORANGE FREE STATE DECEMBER 1981 ISSN 0303-2515 At \ SO Nians A a FEB 17 1982 LIBRARIES ANNALS OF THE SOUTH AFRICAN “MUSEUM CAPE TOWN INSTRUCTIONS TO AUTHORS 1. MATERIAL should be original and not published elsewhere, in whole or in part. 2. LAYOUT should be as follows: (a) Centred masthead to consist of Title: informative but concise, without abbreviations and not including the names of new genera or species Author’s(s’) name(s) Address(es) of author(s) (institution where work was carried out) Number of illustrations (figures, enumerated maps and tables, in this order) (b) Abstract of not more than 200 words, intelligible to the reader without reference to the text (c) Table of contents giving hierarchy of headings and subheadings (d) Introduction : (e) Subject-matter of the paper, divided into sections to correspond with those given in table of contents (f) Summary, if paper is lengthy (g) Acknowledgements (h) References (i) Abbreviations, where these are numerous 3. MANUSCRIPT, to be submitted in triplicate, should be typewritten and neat, double spaced with 2,5 cm margins all round. First lines of paragraphs should be indented. Tables and a list of legends for illustrations should be typed separately, their positions indicated in the text. All pages should be numbered consecutively. Major headings of the paper are centred capitals; first subheadings are shouldered small capitals; second subheadings are shouldered italics; third subheadings are indented, shouldered italics. Further subdivisions should be avoided, as also enumeration (never roman numerals) of headings and abbreviations. Footnotes should be avoided unless they are short and essential. Only generic and specific names should be underlined to indicate italics; all other marking up should be left to editor and publisher. 4. ILLUSTRATIONS should be reducible to a size not exceeding 12 x 18 cm (19 cm including legend); the reduction or enlargement required should be indicated; originals larger than 35 x 47 cm should not be submitted; photographs should be rectangular in shape and final size. A metric scale should appear with all illustrations, otherwise magnification or reduction should be given in the legend; if the latter, then the final reduction or enlargement should be taken into consideration. All illustrations, whether line drawings or photographs, should be termed figures (plates are not printed; half-tones will appear in their proper place in the text) and numbered in a single series. Items of composite figures should be designated by capital letters; lettering of figures is not set in type and should be in lower-case letters. The number of the figure should be lightly marked in pencil on the back of each illustration. 5. REFERENCES cited in text and synonymies should all be included in the list at the end of the paper, using the Harvard System (ibid., idem, loc. cit., op. cit. are not acceptable): (a) Author’s name and year of publication given in text, e.g.: ‘Smith (1969) describes .. .’ ‘Smith (1969: 36, fig. 16) describes...’ “As described (Smith 1969a, 1969b; Jones 1971)’ ‘As described (Haughton & Broom 1927)...’ ‘As described (Haughton et al. 1927)...’ Note: no comma separating name and year Dagination indicated by colon, not p. names of joint authors connected by ampersand - et al. in text for more than two joint authors, but names of all authors given in list of references. (b) Full references at the end of the paper, arranged alphabetically by names, chronologically within each name, with suffixes a, b, etc. to the year for more than one paper by the same author in that year, e.g. Smith (1969a, 19695) and not Smith (1969, 1969a). For books give title in italics, edition, volume number, place of publication, publisher. For journal article give title of article, title of journal in italics (abbreviated according to the World list o, scientific periodicals. 4th ed. London: Butterworths, 1963), series in parentheses, volume number, part number (only if independently paged) in parentheses, pagination (first and last pages of article). Examples (note capitalization and punctuation) BULLOUGH, W. S. 1960. Practical invertebrate anatomy. 2nd ed. London: Macmillan. FISCHER, P.—H. 1948. Données sur la résistance et de le vitalité des mollusques. J. Conch., Paris 88: 100-140. Fiscuer, P.-H., DuvAL, M. & Rarry, A. 1933. Etudes sur les échanges respiratoires des littorines. Archs Zool. exp. gén. 74: 627-634. Konn, A. J. 1960a. Ecological notes on Conus (Mollusca: Gastropoda) in the Trincomalee region of Ceylon. Ann. Mag. nat. Hist. (13) 2: 309-320. Konn, A. J. 19606. Spawning behaviour, egg masses and larval development in Conus from the Indian Ocean. Bull. Bingham oceanogr. Coll. 17 (4): 1-51. THIELE, J. 1910. Mollusca: B. Polyplacophora, Gastropoda marina, Bivalvia. In: SCHULTZE, L. Zoologische und anthropologische Ergebnisse einer Forschungsreise im westlichen und zentralen Siid-Afrika 4: 269-270. Jena: Fischer. Denkschr. med.-naturw. Ges. Jena 16: 269-270. (continued inside back cover) ANNALS OF THE SOUTH AFRICAN MUSEUM ANNALE VAN DIE SUID-AFRIKAANSE MUSEUM Volume 86 }#Band December 1981 Desember Part 1 Deel RELATIVE SIZES OF THE MAXILLARY DECIDUOUS CANINE AND CENTRAL INCISOR TEETH IN THE KALAHARI SAN (BUSHMAN) AND SOUTH AFRICAN NEGRO By Fe] GRINE Cape Town Kaapstad The ANNALS OF THE SOUTH AFRICAN MUSEUM are issued in parts at irregular intervals as material becomes available Obtainable from the South African Museum, P.O. Box 61, Cape Town 8000 Die ANNALE VAN DIE SUID-AFRIKAANSE MUSEUM word uitgegee in dele op ongereelde tye na gelang van die beskikbaarheid van stof Verkrygbaar van die Suid-Afrikaanse Museum, Posbus 61, Kaapstad 8000 OUT OF PRINT/UIT DRUK LAGS SS Glew, AS isso), Ses), 5, 7D), GGL, waa), SOD), &, CD, Ty TOS), 1IC=2, 5) 7) tp.i.)) 15425), 242). 27) 31G23).1326), 33, 4500) EDITOR/REDAKTRISE Ione Rudner Copyright enquiries to the South African Museum Kopieregnavrae aan die Suid-Afrikaanse Museum ISBN 0 86813 026 5 Printed in South Africa by In Suid-Afrika gedruk deur The Rustica Press, Pty., Ltd., Die Rustica-pers, Edms., Bpk., Court Road, Wynberg, Cape Courtweg, Wynberg, Kaap RELATIVE SIZES OF THE MAXILLARY DECIDUOUS CANINE AND CENTRAL INCISOR TEETH IN THE KALAHARI SAN (BUSHMAN) AND SOUTH AFRICAN NEGRO By E> EE. GRINE South African Museum, Cape Town (With | figure and 9 tables) [MS accepted 26 August 1981] ABSTRACT The mean canine-incisor index of the San is low (i.e. the San tend to possess relatively small canines), while the average Negro index is high. The Negro mean index value is rather high compared to values computed for other human groups, while the San index is one of the lowest recorded for any population. The canine-—incisor index, in conjunction with other dental and osteological features, may prove to be useful in the identification of human (par- ticularly immature) remains from African, and especially southern African, archaeological sites. CONTENTS PAGE Introduction : : 2 229 Material é A i ; 230 Methods. : : : 231 Results : : : : D352 Discussion . : : ; 240 Summary . : : ; 242 Acknowledgements . si 248 References . f : é 243 INTRODUCTION In the course of a study of the milk teeth of Japanese and American popu- lations and their offspring, Hanihara (1955) noted a difference in the relative sizes of the maxillary deciduous canines among Japanese, ‘Japanese-American White hybrid’ and ‘Japanese-American Negro hybrid’ samples. With the accumulation of more comparative material, Hanihara (1966) later proposed the relative size of the upper milk canine to be a racial character of the dentition. He compared the size of the maxillary deciduous canine (d°) with that of the maxillary deciduous central incisor (di!) to compute what he termed the ‘canine breadth index’. This index was calculated by the simple formula: MD d Raiviard: where MD d° is the mesiodistal diameter of the canine and MD di! is the corresponding dimension of the incisor. Hanihara observed that in most Caucasoid population samples the mean index value exceeded 105,0 per cent, pad) ] x 100 Ann. S. Afr. Mus. 86 (7), 1981: 229-245, 1 fig., 9 tables. 230 ANNALS OF THE SOUTH AFRICAN MUSEUM while in other groups (e.g. Mongoloid, Negroid, and Australoid) the average index had a value of around 100,0 per cent. A relatively high ‘canine breadth index’ was considered by Hanithara (1967, 1968, 1970) to constitute part of what he called the ‘Caucasoid dental complex’ including also several other features such as a high frequency of the Carabelli trait on the dm? and a low frequency of incisor shovelling. Although Hanihara (1955, 1966, 1967, 1968, 1970) directly recorded indices and calculated average index values from previously published dimen- sional data for several different racial samples, no data are at present available for southern African Negro and Khoisan populations save for the tooth dimensions recorded by Abel (1933). Hanihara (1966: table 5) computed an index from Abel’s data and considered this index as a value for the ‘Bantu’. However, Abel’s data are derived from a mixed sample of Khoi (Hottentot), San (Bushman), and Negro crania, and as such these data are not representative of a homogenous population. (The term ‘Bantu’ refers to the family of languages spoken by southern African Negro peoples and not to a biological group, although the term is often used erroneously in biological contexts.) The present paper records this index of relative d° size for the South African Negro and Kalahari San derived from the author’s unpublished dental measurements on samples drawn from these populations. Note should be made of the term ‘breadth’ as used by Hanihara (1966) in his ‘canine breadth index’. The term ‘breadth’ has been used by various authors in reference to either the mesiodistal or the buccolingual (=labiolingual) diameters of tooth crowns (see Moorrees 1957, Tobias 1967, and Lunt 1969 for complete discussions of this terminological problem). In order to avoid confusion by the use of the term ‘breadth’ in reference to an index calculated from mesiodistal diameters, Hanihara’s ‘canine breadth index’ is probably better referred to by another name. In this paper it will be referred to as the ‘mesiodistal canine-incisor index’. MATERIAL The material examined in this study is drawn from the Khoisan and Negro population groups of southern Africa. The Negro sample comprises crania of South African Bantu-speaking Negroes housed in_ the Raymond A. Dart Collection of Human Skeletons, Department of Anatomy, University of the Witwatersrand, Johannesburg, and also by plaster casts made from the impressions of the teeth and dental arches of living Natal Nguni children. Because of the high incidences of damaged, lost, and diseased teeth in the available Negro samples, only six male and six female subjects were found in which both the di! and d©° could be measured. The term ‘San’ is used in this study to denote a biological population, whereas the name ‘Bushman’ refers more precisely to the ethnological and linguistic practices of these peoples (Jenkins & Tobias 1977). The San sample TEETH IN SAN AND NEGRO 231 is composed entirely of plaster casts made from the impressions of the teeth and jaws of living subjects. The casts were made by J. F. van Reenen and his co-workers on three expeditions to the Kalahari Desert in 1958, 1959, and 1964. Both the dit and d© were measurable on some thirty-four male and twenty-four female specimens. METHODS To measure the diameters of the teeth a sliding vernier caliper equipped with a dial was used. The points of the caliper were tapered and sharpened in order to enable them to be inserted into the interdental embrasures between teeth in the jaw. The mesiodistal crown diameters of the di! and d¢ were recorded to the nearest 0,1 mm. Several different methods have been defined and used in the measurement of mesiodistal crown diameter (Hrdlitka 1923; Campbell 1925; Martin 1928; Remane 1930; Shaw 1931; Pedersen 1949; Selmer-Olsen 1949; Robinson 1956; Moorrees 1957; Korenhof 1960; Goose 1963; Jacobson 1967; Tobias 1967; Lunt 1969). The determination of the mesiodistal diameters of the di! and d° employed here was made according to Tobias’s (1967) definition of this dimen- sion. That is, the mesiodistal diameter was measured as the distance between two parallel lines perpendicular to the mesiodistal axial plane of the tooth and tangential to the most mesial and distal points of the crown. Moreover these points lay along a line parallel to the occlusal plane. This measurement is, accordingly, not dependent upon the position of the tooth in the dental arch and the termini do not necessarily correspond to the points of interstitial contact. Interproximal attrition may have an appreciable effect on the mesiodistal diameter of a tooth crown. In those instances where attritional reduction of tooth material was slight, and the amount of loss could be estimated with a reasonable degree of accuracy, the mesiodistal measurement was corrected so that it approximated the unworn diameter. For teeth that were heavily worn, and for which mensurable correction was more a ‘guesstimate’ than a reason- able estimate, a dimension was not recorded. The mesiodistal diameters of the di! and d© were recorded and the canine- incisor index was calculated separately for each individual according to the formula given above. Only a single incisor and a single canine (usually the left? were measured for each individual. The data was processed by a shortened version of the OSIRIS programme, and the computation of statistics was performed using the SPSS programmes package (Nie et al. 1975). Except for the indices recorded by Hanihara for sexually pooled samples of Japanese, Amerindians (Pima), Alaskan Eskimos, American Caucasoids, American Negroes, ‘Japanese-American White hybrids’ and ‘Japanese- American Negro hybrids’ (1966: table 2), all the indices for comparative populations used in this paper were calculated from reported means of mesio- 232 ANNALS OF THE SOUTH AFRICAN MUSEUM distal diameters of the di! and d°. It should be noted that a mean index and an index of means computed for a single population sample will not necessarily be the same. For example, the canine-incisor index of San males calculated as an average of individual indicial values (a mean index) 1s 97,8 per cent (Table 1), while the index computed from the mean di! and d° dimensions (an index of means) 1s some 98,2 per cent (Table 3). Nevertheless, the different values obtained by these two methods of calculation appear to be similar enough to permit comparison of the present results with the indices computed from the mean crown diameters recorded for other population groups. RESULTS The canine-incisor indices obtained for the Kalahari San and South African Negro samples are recorded in Table |. It is immediately apparent that on average the San possesses a relatively smaller maxillary deciduous canine TABLE | Canine-incisor indices of the Kalahari San and South African Negro. Population sample Sample range Sanpmale a: a. sa. wt ee 90,0-113,3 Samulemales y sao 2 ec ote 89,2-109,5 San male-- female << . .-. 89,2-113,3 INGprOmmaless cao Sy uae 7 eee 100,0-122,4 INGgRO Teme ne 91,6-113,6 Negro male + female . . . 91,6-122,4 N—sample size; X —sample mean; SE~—standard error of the mean; SD—standard deviation; CV—coefficient of variation (SD/X(100)). than does the South African Negro. All San sample means fall below 100,0 per cent, while all Negro means fall above 100,0 per cent. Individual index values for all samples, except the Negro male, range from below to above 100,0 per cent. Despite the small sizes of the Negro male and female samples, F analyses of variance revealed no apparent significant differences in the index value distributions amongst these and the other samples considered here. Statistical testing of the means of the various samples, notwithstanding the small sizes of some, was considered justified. The results of these comparisons are given in Table 2. The mean index values of the San male and female samples are nearly the same (Table 1) and, as expected, this slight difference was found to be not statistically significant. The mean index value of the Negro male sample is somewhat higher than that of the Negro female sample (Table 1), but this mean difference was also found to be not statistically significant. Thus no apparent sexual differences are shown by the San and Negro samples in the TEETH IN SAN AND NEGRO 235 TABLE 2 Statistical comparison of mean canine-incisor index values amongst the Kalahari San and South African Negro samples. Sample comparisons Significance San male—San female Negro male—Negro female San male—Negro male . . . p < 0,005 San female—Negro female ... — San—Negro (pooled samples) . p < 0,001 df—degrees of freedom. TABLE 3 Canine-incisor indices computed for different populations. Males. Population (mm) | (mm) Reference MONGOLOID a J2GANESE: 1S ee 6,63 6,75 101,8 Ono 1960 JADEMESS 2 Reece 6,70 6,70 100,0 Hanihara 1976 Amenndian (Pima) . . . 6,86 UMS 104,2 Hanihara 1976 CAUCASOID | RSUUNETENOE oe 6,72 6,99 104,0 Boboc 1965 Byte , |) iin 6,46 6,78 105,0 Clinch 1963 SWWEGISN: 5 Soll tn 6,60 7,04 106,7 Seipel 1946 INOntheAmernican . °. . .. 6,40 6,76 105,9 Black 1978 INonthneAmentcan 6. 6,40 6,80 106,3 Hanihara 1976 INOKtheAMenIcam ... . .. 6,55 6,88 105,0 Moorrees 1959 North American AUSTRALOID Australian Aborigine . ... i385 7,41 Australian Aborigine... 7,40 ee Australian Aborigine. ... V3) UBS Naisoi Bee: NEGROID AAMenMICAMUNESTO. . . =. ; 6,03 6,86 South African Negro. ... South African Negro. ... 6,47 7,08 KHOISANOID Kalahari San Kalahari San MIXED SAMPLE Tristanite 105,4 Meredith & Knott 1968 100,8 Margetts & Brown 1978 101,6 Barrett, Brown & Luke 1963 100,6 Hanihara 1976 99,0 Bailit, De Witt & Leigh 1968 Hanihara 1976 Present study? Grine unpublished? + Present study? Grine unpublished? + Thomsen 1955 + —mean of individual index values (mean index); ++ —index calculated from mean dimensions of di! and d° (index of means). canine-incisor index. Hanihara (1955, 1966) found no significant sexual differ- ences in the index values of any of the populations measured by him. The Negro male displays, on average, a higher canine-incisor index than does the San male, and comparison of these mean values indicated this differ- 234 ANNALS OF THE SOUTH AFRICAN MUSEUM ence to be statistically significant (Table 2: ‘t? = —3,27; p < 0,005). The mean indicial value of the Negro female is higher than that of the San female; but comparison of these values indicated the difference to be not statistically significant, where a value of p < 0,05 was considered to represent significance. As would be expected from the foregoing the mean index value of the sexually pooled Negro sample is higher than the average value obtained for the pooled San sample (Table 1). Comparison of the pooled means (Table 2) indicated the Negro value to be significantly higher than that of the San (‘t? = —3,50; p= O00): The mean canine-incisor indices of the San and Negro male samples are compared with indices computed for male samples of other population groups in Table 3. It appears that while Caucasoid males display a relatively high index with values ranging from 104,0 to 106,7 per cent, Australoid males show a relatively low index with values ranging from 99,0 to 101,6 per cent. Males of the three Mongoloid populations recorded (Table 3) appear to show index values intermediate between the Australoid and Caucasoid values. TABLE 4 Canine-incisor indices computed for different populations. Females. MD di!| MD d& | Index (mm) | (mm) We MONGOLOID JAPANESE ak Ae Re ae mere 6,57 JapameSsea an) = Reference Ono 1960 Hanihara 1976 Amerindian (Pima) . ... 6,82 Hanihara 1976 CAUCASOID Rumanian a ae 5k? 6,52 Boboc 1965 BiitiSht 52 Pyseh ies se Rei 6,31 6,61 Clinch 1963 SwediSinge: ka 27 08 Us tee, 6,56 6,93 Seipel 1946 INoneheAinenicane yo) ee 6,52 6,66 Black 1978 North American, 998)5) 2a ae 6,40 6,76 Hanihara 1976 North American: 2 2) 5) 6,44 6,67 Moorrees 1959 INonéheAmencanhe = | ole ae 37 Meredith & Knott 1968 AUSTRALOID Australian Aborigine . . . 7,20 Margetts & Brown 1978 Australian Aborigine. . . V2 Barrett, Brown & Luke 1963 Hanihara 1976 Bailit, De Witt & Leigh 1968 Australian Aborigine. . . UBS IN@ISOIT | (i eee? el 8 Ta oe 6,71 NEGROID North American Negro . . 6,54 South African Negro . : South African Negro. . . 6,58 KHOISANOID Kalahari San Kalahari San MIXED SAMPLE Tristanite Hanihara 1976 Present studyt Grine unpublished? + Present studyt Grine unpublished? t Thomsen 1955 + —mean of individual index values (mean index); ++ —index calculated from mean dimensions of di? and d°¢ (index of means). TEETH IN SAN AND NEGRO DBS Interestingly, the American Negro and South African Negro males possess the highest canine-incisor indices of any of the population groups compared, with values of 113,8 and 107,0 per cent (or 109,4 depending upon the mode of calculation) respectively. The index value of the San male, 97,8 per cent (or 98,2 depending upon the mode of calculation) is lower than that of any other male sample save the Tristanites. Comparison of the mean index values recorded for the San and Negro female samples with the indices of means calculated for other female popu- lations (Table 4) reveals a pattern of distribution similar to that described for males. The mean index of the South African Negro female (102,9°%) falls within the range shown by Caucasoid female samples, whereas the South African Negro female index of means has a value of 100,8 per cent. The latter value is almost identical to that of the mixed Tristanite population, 1s lower than any Caucasoid or Mongoloid female value recorded, but 1s slightly larger than the values obtained for Australoid female samples. Only the Naisoi female index of means is lower than the mean index of the San female (98,0°,), but the index value obtained from the mean tooth dimensions of the San female (96,9 °, ) is lower than that recorded for any other population. Generally speaking, it appears that for most groups the mean female index value is somewhat lower than the average male index value. The canine-incisor index of the sexually pooled South African Negro sample (105,0 or 104,3°. depending upon the mode of calculation) is higher than that recorded for any Mongoloid or Australoid population, and falls within the ranges obtained for Caucasoids and established for other Negro populations (Table 5). The South African Negro index is considerably higher than that calculated for a sample of Liberian Negroes. It is also higher than the two index values recorded by Hanihara (1955, 1966) for ‘Japanese-American Negro hybrids’, and the index value calculated from Abel’s (1933) data for a mixed sample of San, Khoi, and Negro individuals. The mean index, as well as the index of the means calculated for the South African Negro, are nearly the Same as the mean index value recorded by Hanihara (1966, 1970) for the American Negro, but they are somewhat lower than the index values for American Negroes recorded elsewhere by Hanihara (1967) and computed from his 1976 data. The San indicial value (either 97,9 or 97,6°.) is lower than any recorded for a Caucasoid sample (Table 5). The San values are lower also than most obtained for Mongoloid populations, although they are approached by the values calculated for the Ainu (Hanihara 1970) and the Japanese sample measured by Arai (1937). The San values bracket the index value obtained by Hanthara (1955) for a Japanese sample. The San values are lower than those obtained for some Australoid populations, but are almost the same as the value calculated for the Naisoi, while the Australian Aboriginal index calculated from Campbell’s (1925) data is lower. The Tristanite index value is also close to the San values. The San index values are lower than any recorded for a 236 ANNALS OF THE SOUTH AFRICAN MUSEUM TABLE 5 Canine-incisor indices reported or calculated for different populations. Sexually pooled samples. MD d°¢ | Index Population MONGOLOID JAMWANESE. sea ee ete a eas Hanihara 1955 NapaneSee 45 we ee ees be 6,60 6,71 101,7 Ono 1960* JaWAMESE: 45, We ew en ee 6,64 6,60 99,4 Arai 1937 Japanese 101,0 Hanihara 1966, 1970 JADAMESE? Fee oe era ees 101,5 Hanihara 1967 ValpaneSe: "gar peeve eae es 6,62 6,66 100,6 Hanihara 1976* Ainu Sees ec a ec Seen see? 99,5 Hanihara 1970 Amerindian (Pima) . . . 103,3 Hanihara 1966, 1970 Amerindian (Pima) . .. . 6,84 7,08 103,5 Hanihara 1976 Amerindian (GKAH). .. . 6,28 6,37 101,4 Sciulli 1977 Amerindian (PH) . : 6,42 6,44 100,3 Sciulli 1977 Alaskan Eskimo 100,3 Hanihara 1966, 1970 Aleut Eskimo 100,0 Hanihara 1966 CAUCASOID Rumanian 6,62 6,86 103,6 Boboc 1965* Belgian medieval 6,33 6,85 108,2 Mydlarz 1964 European 6,10 6,70 109,8 Brabant 1965 Polish 6,40 6,90 107,8 Szlachetko 1959 British 6,30 6,60 104,8 Dolamore 1908 British 6,39 6,70 104,9 Clinch 1963 Swedish . 6,58 6,99 106,2 Seipel 1946* Aran Islander . 6,30 6,80 107,9 Dockrell 1956 North American 6,50 7,00 OVE Black 1897 North American 6,46 6,72 104,0 Black 1978* North American 6,40 6,78 105,9 Hanihara 1976* North American 6,50 6,90 106,2 Jones 1947 North American 6,50 6,78 104,3 Moorrees 1959* North American 6,55 6,82 104,1 Kramer & Ireland 1959 North American 108,2 Hanihara 1966, 1970 North American 106,3 Hanihara 1967 North American 6,39 6,77 106,0 Meredith & Knott 1968 North American 6,40 6,80 106,3 Raak 1950 AUSTRALOID Australian Aborigine... 7,28 3! 100,4 Margetts & Brown 1978* Australian Aborigine . . . 35 Ye 101,0 Barrett, Brown & Luke 1963 Australian Aborigine. .. 7,30 Wo 322 100,3 Hanihara 1976* Australian Aborigine. .. 7,80 7,50 96,2 Campbell 1925 INAISO li tees ak were se 8 Oni 6,64 98,1 Bailit, De Witt & Leigh 1968’ NEGROID AmericanNesro) % a8) 2 - 104,9 Hanihara 1966, 1970 AIMEGICAT INGSTOM sn) 5) menue 107,8 Hanihara 1967 American Neeior 4 eas |: 6,29 6,88 109,4 Hanihara 1976* EibenanyNecron 7 an 6,85 6,79 99,1 Moss & Chase 1966 South African Negro. .. 105,0 Present studyt South African Negro. . . 6,53 6,59 104,3 Grine unpublishedt + KHOISANOID KalaharcSan' 22°. 0. eae 97,9 Present studyt Kalahariesan’.. 8, aa. 6,75 6,59 97,6 Grine unpublished? 7 TEETH IN SAN AND NEGRO 237 TABLE 5 (continued) MD di!| MD d¢ | Index Rewerence (mm) | (mm) WA | MIXED SAMPLES Japanese-Caucasoid . . . Hanihara 1955 Japanese-Caucasoid . . . Hanihara 1966 WapaMnese-Neero . . . . Hanihara 1955 Japanese-Neeto - . « = : Hanihara 1966 San-Khoi-Negro .. .. . 6,60 6,60 Abel 1933 Ministamite, 2 . . =. . . 6,51 6,44 Thomsen 1955 In those instances where no MD diameters for di! and d° are given, the index value was reported directly. *—index values calculated from averaged mean diameters of di’ and d° and are not the averages of male and female indices given in Tables 3 and 4. +—mean of individual index values (mean index); +{—index calculated from the mean dimensions of di! and d°¢ (index of means). Negro population, although they are approximated by the values calculated for Liberian Negroes and a sample of ‘Japanese-American Negro hybrids’ examined by Hanihara (1955). Thus the mean mesiodistal canine-incisor index value of the South African Negro appears to be relatively high, while that of the San seems to be relatively low compared with the values obtained for other human population samples. Hanihara (1955, 1966) recorded the frequencies of individuals from several populations in whom the d° was larger than the di’ (index values equal to or greater than 100,1°%) and in whom the d° was smaller than or equal to the size of the di! (index values of x—100,0°%). He (1966) recorded that the American Caucasoid sample had a higher frequency of individuals in whom the d° was mesiodistally larger than the di! than did any of the other population samples examined by him. The percentage frequencies of individuals with relatively small canines (index values of x-100,0°%) and relatively large canines (index values of 100,1—x) in the South African Negro and San samples are recorded in Table 6. While no significant sexual differences in frequency distributions within the San and Negro samples were found (Table 7), it is readily apparent that, whereas the TABLE 6 Numbers and frequencies of individuals in the South African Negro and Kalahari San samples in whom the deciduous canine is smaller than, and/or equal to, or larger than the central deciduous incisor. Index value x-100,0% Index value 100,1-x % Population sample San male San female . oe San—male + female . Negro male : INesro female... Negro—male + female 238 ANNALS OF THE SOUTH AFRICAN MUSEUM TABLE 7 Chi-square evaluation of distribution of index values above and below 100,0 per cent amongst the San and South African Negro. Sample comparisons significance San male—San female. . . . — Negro male— Negro female San male—Negro male . . . p < 0,01 San female—Negro female . . — San—Negro (pooled samples) . p < 0,001 TABLE 8 -Comparison of frequencies of Kalahari San and South African Negro individuals with large or small canine-incisor indices with frequencies reported for other populations. Sexually pooled samples. 5 Index value 100,1—x DAY Hanihara 19 Hanihara 1966 Hanihara 1966 Hanihara 1966 Hanihara 1966 Hanihara 1966 Present study Amerindian (Pima) . Eskimo (Alaskan) American—Caucasoid American Negro South African Negro Kalahari San Present study Japanese—Caucasoid Hanihara 1955 Japanese—Negro Soc: 13 Hanihara 1955 Numbers of individuals for other populations calculated from data presented by Hanihara (1966: table 4). majority of San individuals have relatively small deciduous canines, the majority of Negro individuals have relatively large canines. Chi-square evaluations of the frequencies shown by the San and Negro samples revealed the difference between San and Negro males to be significant (x? = 7,25; p < 0,01), while the difference among the San and Negro females appeared to be not statistically significant. The distributional differences among the sexually pooled San and Negro samples were found to be highly significant (x? = 9,74; p < 0,001). While some 75,0 per cent of Negro individuals examined have maxillary decidu- ous canines that are mesiodistally larger than the maxillary deciduous central incisors, almost three-quarters of San individuals have incisors that are not as large as, or larger than, the canines. The frequency distributions of relative canine size in the sexually pooled San and Negro samples are compared in Table 8 with these distributions for other populations as recorded by Hanihara (1955, 1966). It would appear that in Mongoloid peoples the percentage frequency of individuals with relatively large canines tends to be consistently lower than the corresponding frequencies TEETH IN SAN AND NEGRO 239 of Negro individuals. The Caucasoid sample studied by Hanihara (1966) has the highest frequency of individuals in whom the d° is larger than the di’. Comparison of the South African Negro frequency distributions with those of other populations (Table 9) indicated the differences to be not statistically significant, except in the case of Hanihara’s (1955) Japanese sample. However, the frequency distributions of Hanihara’s (1955, 1966) two Japanese samples are also significantly different. Thus, the percentage frequency of South African Negro individuals with relatively large canines is comparable to that in the other population groups sampled. The San, however, shows a distributional pattern which is markedly different from that found in any other population, save Hanihara’s (1955) Japanese sample, for which these frequencies have been recorded (Table 8). Comparison of the distribution frequencies encountered in the San sample with the frequencies recorded for other populations indicated the San distri- bution to be significantly different in each case except for Hanihara’s (1955) Japanese sample (Table 9). The percentage of San individuals in whom the d° TABLE 9 Chi-square evaluation of distribution of individual index values above and below 100,0 per cent amongst the San, South African Negro, and other populations. Sexually pooled samples. Comparative population x2 significance KALAHARI SAN COMPARISONS Japanese (Hanihara 1955)... 0,48 — Japanese (Hanihara 1966). . . 11,14 p < 0,001 Amenindianio se ns 18,55 p < 0,001 FISISIMORee (GRR os Pe 6,70 () KOLO | @aqucasordis 9s. sack ao: 21,18 p < 0,001 EN INSHCIO ING TO) 5 10,26 p < 0,01 Japanese-Caucasoid . . . . 12,19 p < 0,001 JEVORINESSSINISA) 5 5 6 6 6 4,58 p < 0,05 SOUTH AFRICAN NEGRO COMPARISONS Japanese (Hanihara 1955)... 13,08 p < 0,001 Japanese (Hanihara 1966). . . 1,92 — AMmennGgiane Geri kerr Ae 6 0,64 os IS KAUN Og My tale ak ida wR oe 0,35 — CaMeaSOrd ine wes BG eS My ee 1,00 _ JNO INGYR) 5 6 » 2 4 0,08 — Japanese-Caucasoid . . . . 0,85 _— Japanese-Negro .... . 1,78 — is larger than the di! is markedly smaller than this frequency recorded for any other population group, except the Japanese studied by Hanihara (1955). Apart from the fact that the Japanese sample examined by Hanihara in 1955 240 ANNALS OF THE SOUTH AFRICAN MUSEUM is about half as large as that studied by him in 1966, Hanihara has not provided an explanation for the significant differences in frequency distributions between these two samples. DISCUSSION Hanihara (1966, 1967, 1968, 1970) proposed that a relatively high canine— incisor index (i.e. values of about 105,0°% and over) was distinctive of Cauca- soid peoples and formed part of the ‘Caucasoid dental complex’. Comparisons of canine-incisor indices for a larger series of population samples appear to support Hanihara’s statements that Caucasoids tend to have higher mean index values than most other groups. Australoid peoples generally have lower mean indicial values, while Mongoloid populations tend to show intermediate values. The canine-incisor index of the South African Negro 1s relatively high. In all three sets of comparisons (male, female, and sexually pooled samples) the mean Negro values fell within the observed Caucasoid ranges. The canine-incisor index of the Kalahari San is, on average, rather low. In all three sets of comparisons the mean San values were found to be among the lowest computed for all populations. Index values obtained for the American Negro tend to be high, falling within or even above observed Caucasoid mean ranges, and Hanihara (1967, 1970) has ascribed this to genetic admixture with American Caucasoids. Interestingly, while the canine—incisor index values of the South African Negro are also high, falling within the observed Caucasoid mean ranges but generally below the mean American Negro values, the index computed for the Liberian Negro is considerably lower than that shown by the South African Negro sample. Unfortunately, the data presented by Moss & Chase (1966) for Liberian Negroes and those given here for South African Negroes are the only measure- ments available at present for African Negro populations. More data will have to be accumulated for other African Negro samples before the rather marked difference in index values between the Liberian and South African Negroes can be explained adequately. Perhaps the most striking features to emerge from this study are (1) the marked difference between the mean canine-incisor index values shown by the San and South African Negro populations, and (ii) the marked difference in the frequency distributions of individual index values among these samples. The tendency towards a low mesiodistal canine-incisor index may be added to the list of osteological (De Villiers 1968; Grine 1979), somatic (Drury & Drennan 1926; Tobias 1955-1956, 1960, 1961, 1966; De Villiers 1961; Singer & Weiner 1963; Weiner et al. 1964), dental (Van Reenen 1961, 1964, 1966; Van Reenen & Dreyer 1963; Grine 1978, 1981), and genetic (Tobias 1972; Nurse & Jenkins 1977) features that characterize the San and serve to differen- tiate them from other African Negroid groups. 24] TEETH IN SAN AND NEGRO (dS I + X URS) s’cOT PUR (GS I — X O1BON) Z'L6 USEMJ9q PUB MOTIq ‘QAOQE SON|LA XOPUI YIM S[VNPIAIPU! OIBINY puke ues JO SaIdUaNbaJ JUISoIdal SOINSY 9dVjUIIIIg “XIPUI JOSIOUI—OUIUBDS [LISIPOISAW 9Y} JO UOIBIAIP Pplepueys suo SiG + SUBSLU PUB SUBaLU ‘sosUeI sj]dWes OIBZONY UBdIIJW YINOS pue ues psjood Ajjenxas 9y} JO UO!}eJUSSaIda1 DeWURIZRIG SANIVA XSONI OCI ll Oll SOl OO| G6 06 G8 S005 els Ve ‘8 —> een ara FAAS ol, [G-———> gE0! ZL6 ‘| ‘314 ODay UPS olan ues DAD ANNALS OF THE SOUTH AFRICAN MUSEUM The mesiodistal canine-incisor index may serve as a useful tool to aid in the identification of human population samples from African, and especially southern African, archaeological sites. The racial identification of immature remains from such contexts is often difficult (Grine 1979) and it is in this regard that the canine-incisor index would appear to be of particular use. The dis- criminant reliability of this index, as with all other metrical and non-metrical characters of the dentition and skeleton, would be enhanced in circumstances where the population sample in question is of at least moderate size. Unfortu- nately, however, human remains from archaeological sites are often few in number. In the identification of individual specimens, comparative sample ranges of a measurement are commonly used. There is a considerable degree of overlap in individual canine-incisor index values among the San and South African - Negro (Figure |, Table 1). Thus according to the data derived from the present samples of these two populations, a child with an index value of between 91,6 and 113,3 per cent could belong to either group. However, even if only one standard deviation on either side of the mean is used rather than the entire sample range, the utility of the mesiodistal canine—incisor index is noticeably increased. | For the present samples the upper extreme of one standard deviation of the San mean 1s a value of 103,75 per cent and the lower extreme of one standard deviation of the Negro mean is 97,21 (Fig. 1). Some 51,7 per cent of individuals comprising the San sample show index values below 97,21, while 50,0 per cent of Negro individuals examined have indicial values above 103,75. SUMMARY The relative mesiodistal diameters of the d¢ and di! in samples of Kalahari San and South African Negro populations are investigated. The San tends to have a low canine-incisor index (i.e. the canines tend to be relatively small), while the Negro generally possesses a high canine-incisor index (i.e. the canines tend to be relatively large). Generally, Caucasoid samples have relatively high index values (Hanihara 1966, 1967, 1968, 1970), Australoid populations show relatively low values, and the indices of Mongoloid populations tend to be intermediate between these two extremes. The South African and American Negro index values but not the Liberian Negro index value are comparable to those displayed by Caucasoid samples. The mean San index is among the lowest computed for any human population. The mean index values as well as the frequency distributions of individual index values are markedly different between the San and South African Negro. These differences are most apparent between the male and sexually pooled samples. Used in conjunction with other dental and osteological features, the mesio- distal canine-incisor index may prove to be a useful tool in the identification TEETH IN SAN AND NEGRO 243 of human population samples from African, and especially southern African, archaeological sites. This index would appear to be particularly useful in the analysis of immature human remains. ACKNOWLEDGEMENTS I thank Prof. J. F. van Reenen for permission to examine the casts of the Kalahari San, Prof. P. V. Tobias for permission to examine material in the Raymond A. Dart skeletal collection, and Dr C. Catterick for making casts of Nata! Nguni children available to me. My thanks go to Profs P. V. Tobias, H. de Villiers, and R. Klein for reading this paper and for their constructive comments. Mrs P. Eedes typed the manuscript. REFERENCES ABEL, W. 1933. Zahne und Kiefer in ihrem Wechselbeziehungen bei Buschmannern, Hotten- totten, Negern und deren Bastarden. Z. Morph. Anthrop. 31: 314-361. ARAI, T. 1937. Intercorrelations between mesiodistal and buccolingual diameters in the deciduous and permanent teeth. J. Jap. dent. Soc. 77: 176-185. BaiLit, H. L., DE Witt, S. J. & LEIGH, R. A. 1968. The size and morphology of the Naisoi dentition. Am. J. phys. Anthrop. 28: 271-288. BARRETT, M., BROWN, T. & LUKE, J. 1963. Dental observations on Australian Aborigines: mesiodistal crown diameters of deciduous teeth. Aust. dent. J. 8: 299-302. BLACK, G. V. 1897. Descriptive anatomy of the human teeth. Philadelphia: S. S. White Dental Mfg. BLACK, T. K. 1978. Sexual dimorphism in the tooth-crown diameters of the deciduous teeth. Am. J. phys. Anthrop. 48: 77-82. Bosoc, G. 1965. Studiu statistic si citera consideratii asupra latimii dintilor temorari si definitivi. Stomatologia 12: 345-354. BRABANT, H. 1965. Observations sur |’evolution de la denture temporaire humaine en Europe occidentale. Bull. Grpmt int. Rech. scient. Stomat. 8: 235-302. CAMPBELL, T. D. 1925. Dentition and palate of the Australian Aboriginal. Publs Keith Sheridan Fdn med. Res. 1: 1-123. CLINCH, L. M. 1963. A longitudinal study of the mesiodistal crown diameters of the deciduous teeth and their permanent successors. Europ. orthod. Soc. Rep. 39: 202-215. De Vicuiers, H. 1961. The tablier and steatopygia in Kalahari Bushwomen. S. Afr. J. Sci. 57: 223-227. De ViLuErs, H. 1968. The skull of the South African Negro: a biometrical and morphological study. Johannesburg: Witwatersrand University Press. DOocKRELL, R. B. 1956. Tooth size in Irish (Aran Island) families. Europ. orthod. Soc. Rep. 32: 200-216. DoLamore, W. H. 1908. The relation of the deciduous to the permanent teeth. Gaz. R. dent. Hosp. 1908: 144-156. Drury, D. H. & DRENNAN, M. R. 1926. The pudendal parts of the South African Bush race. S. Afr. med. J. 22: 113-117. Goose, D. H. 1963. Dental measurement: an assessment of its value in anthropological studies. Jn: BROTHWELL, D. R., ed. Dental anthropology: 125-148. London: Pergamon. GRINE, F. E. 1978. Occlusal morphology of the lower molars of the South African Negro and San (Bushman). J. Anat. 126: 667. GRINE, F. E. 1979. Report on two infant skeletons and an adult metatarsal excavated from Diana’s Vow. Occ. Pap. natn. Mus. Rhod. (A) 4: 141-145. GRINE, F. E. 1981. Occlusal morphology of the mandibular permanent molars of the South African Negro and the Kalahari San (Bushman). Ann. S. Afr. Mus. 86: 157-215. HANIHARA, K. 1955. Studies on the deciduous dentition of the Japanese and the Japanese— American hybrids. II. Deciduous canines. J. anthrop. Soc. Nippon 64: 63-82. 244 ANNALS OF THE SOUTH AFRICAN MUSEUM HANIHARA, K. 1966. Relative size of the deciduous upper canine —a new racial character in the dentition. J. anthrop. Soc. Nippon 74: 9-17. HANIHARA, K. 1967. Racial characters in the dentition. J. dent. Res. 46: 923-926. HANIHARA, K. 1968. Morphological pattern of the deciduous dentition in the Japanese— American hybrids. J. anthrop. Soc. Nippon 76: 114-121. HANIHARA, K. 1970. Mongoloid dental complex in the deciduous dentition with special reference to the dentition of the Ainu. J. anthrop. Soc. Nippon 78: 3-17. HANIHARA, K. 1976. Statistical and comparative studies of the Australian aboriginal dentition. Bull. Univ. Mus. Tokyo 11: 1-57. Hrb.ickA, A. 1923. Variation in the dimensions of lower molars in man and anthropoid apes. Am. J. phys. Anthrop. 16: 123-166. JACOBSON, A. 1967. The Bantu dentition: a morphological and metrical study of the teeth, the jaws and the bony palate of several large groups of South African Bantu-speaking Negroids. Unpublished Ph.D. thesis, University of the Witwatersrand. JENKINS, T. & TosiAs, P. V. 1977. Nomenclature of population groups in southern Africa. Afr. Stud. 36: 49-55. Jones, H. G. 1947. The primary dentition in Homo sapiens and the search for primitive features. Am. J. phys. Anthrop. 5: 251-282. KorENHOF, C. A. W. 1960. Morphogenetical aspects of the human upper molar. Utrecht: Uitgeversmaatschappij Neerlandia. KRAMER, W. S. & IRELAND, R. 1959. Measurements of the primary teeth. J. Dent. Child. 26: 252-261. Lunt, D. A. 1969. An odontometric study of mediaeval Danes. Acta odont. Scand. 27 (suppl. 55): 1-173. Marcetts, B. & Brown, T. 1978. Crown diameters of the deciduous teeth in Australian aboriginals. Am. J. phys. Anthrop. 48: 493-S02. MartTIN, R. 1928. Lehrbuch der Anthropologie. Part 11. 2nd ed. Jena: Gustav Fischer. MEREDITH, H. V. & Knott, V. B. 1968. Coronal breadth of human primary anterior teeth. Am. J. phys. Anthrop. 28: 49-64. Moorregs, C. F. A. 1957. The Aleut dentition. Cambridge (Mass.): Harvard University Press. Moorregs, C. F. A. 1959. The dentition of the growing child: a longitudinal study of dental development between 3 and 18 years of age. Cambridge (Mass.): Harvard University Press. Moss, M. L. & CHAsE, P. S. 1966. Morphology of Liberian Negro deciduous teeth. I. Odonto- metry. Am. J. phys. Anthrop. 24: 215-230. MyYDLARZ, A. 1964. Observations sur les dimensions de dents temporaires d’age Médieéval. Bull. Grpmt int. Rech. scient. Stomat. 7: 121-141. Nig, N. H., Hut, C. H., JENKINS, J. G., STEINBRENNER, K. & BENT, D. H. 1975. Statistical package for the social sciences. 2nd ed. New York: McGraw-Hill. Nurse, G. T. & JENKINS, T. 1977. Health and the hunter-gatherer: biomedical studies on the hunting and gathering populations of southern Africa. Monographs in human genetics 8: 1-126. Basle: S. Karger. Ono, H. 1960. Mesiodistal diameters of primary and permanent teeth and their correlation in the arch. J. Jap. stomat. Soc. 27; 221-234. PEDERSEN, P. O. 1949. The East Greenland Eskimo dentition. Copenhagen: C. A. Reitzels. RAAK, K. D. 1950. The mesio-distal diameter of the deciduous teeth. Unpublished M.S. thesis, University of lowa, Iowa City. REMANE, A. 1930. Zur Messtechnik der Primatanzahne. Jn: ABERHALDEN, E., ed. Handbuch der biologischen Arbeitsmethoden 7: 609-635. RosInson, J. T. 1956. The dentition of the Australopithecinae. Mem. Transv. Mus. 9: 1-179. ScIuLLI, P. W. 1977. A descriptive and comparative study of the deciduous dentition of prehistoric Ohio Valley Amerindians. Am. J. phys. Anthrop. 47: 71-80. SEIPEL, C. 1946. Variation in tooth position: a metric study of variation and adaptation in the deciduous and permanent dentitions. Swed. dent. J. 39 (suppl.): 1-176. SELMER-OLSEN, R. 1949. An odontological study of the Norwegian Lapps. Skrif. norske Vidensk-Akad. mat.-nat. Kl. 3: 1-167. SHAw, J. C. M. 1931. The teeth, the bony palate and the mandible in Bantu races in South Africa. London: J. Bale, Sons and Danielsson. SINGER, R. & WEINER, J. S. 1963. Biological aspects of some indigenous African populations. S West. J. Anthrop. 19: 168-176. TEETH IN SAN AND NEGRO 245 SZLACHETKO, K. 1959. Investigations on the morphology of the human deciduous dentition. Acta Fac. Rerum nat. Univ. comen., Bratisl. 3: 247-279. THOMSEN, S. 1955. Dental morphology and occlusion in the people of Tristan da Cunha. Oslo: Norske Videnkaps-Akademi. Tosias, P. V. 1955-56. Les Bochimans Auen et Naron de Ghanzi. Contribution a l’étude des ‘Anciens Jaunes’ sud-africains. I-IV. Anthropologie 59: 235-252, 429-461, 60: 22-52, 268-289. Tosias, P. V. 1960. Bushman hunter—gatherers: a study in human ecology. Jn: Davies, D. H. S.., ed. Ecological studies in southern Africa: 67-86. The Hague: W. Junk. Tosias, P. V. 1961. Fingerprints and palmar prints of Kalahari Bushmen. S. Afr. J. Sci. 57: 333-345. Tosias, P. V. 1966. The peoples of Africa south of the Sahara. Jn: BAKER, D. T. and WEINER, J. S., eds. The biology of human adaptability: 111-200. Oxford: Clarendon Press. Tosias, P. V. 1967. The cranium and maxillary dentition of Australopithecus (Zinjanthropus) boisei. Olduvai Gorge 2. Cambridge: Cambridge University Press. Tosias, P. V. 1972. Recent human biological studies in southern Africa, with special reference to the Negroes and Khoisans. Trans. R. Soc. S. Afr. 40: 109-133. VAN REENEN, J. F. 1961. The use of the concept of tooth material as an indication of tooth size in a group of Kalahari Bushmen. S. Afr. J. Sci. 57: 347-352. VAN REENEN, J. F. 1964. Dentition, jaws and palate of the Kalahari Bushmen. J. dent. Assoc. S. Afr. 19: 1-37. VAN REENEN, J. F. 1966. Dental features of a low caries primitive population. J. dent. Res. 45 (suppl. 1): 703-713. VAN REENEN, J. F. & Dreyer, C. J. 1963. The relationship between tooth material and the supporting structures of the teeth in a group of Kalahari Bushmen: a comparative study. J. dent. Assoc. S. Afr. 18: 474-479. WEINER, J. S., AINSWORTH HARRISON, G., SINGER, R., HARRIS, R. & Jopp, W. 1964. Skin colour in southern Africa. Hum. Biol. 36: 294-307. 6. SYSTEMATIC papers must conform to the International code of zoological nomenclature (particularly Articles 22 and 51). Names of new taxa, combinations, synonyms, etc., when used for the first time, must be followed by the appropriate Latin (not English) abbreviation, e.g. gen. nov., sp. nov., comb. nov., syn. nov., etc. An author’s name when cited must follow the name of the taxon without intervening punctuation and not be abbreviated; if the year is added, a comma must separate author’s name and year. The author’s name (and date, if cited) must be placed in parentheses if a species or subspecies is transferred from its original genus. The name of a subsequent user of a scientific name must be separated from the scientific name by a colon. Synonymy arrangement should be according to chronology of names, i.e. all published scientific names by which the species previously has been designated are listed in chronological order, with all references to that name following in chronological order, e.g.: Family Nuculanidae Nuculana (Lembulus) bicuspidata (Gould, 1845) Figs 14-15A Nucula (Leda) bicuspidata Gould, 1845: 37. Leda plicifera A. Adams, 1856: 50. Laeda bicuspidata Hanley, 1859: 118, pl. 228 (fig. 73). Sowerby, 1871: pl. 2 (fig. 8a—b). Nucula largillierti Philippi, 1861: 87. Leda bicuspidata: Nicklés, 1950: 163, fig. 301; 1955: 110. Barnard, 1964: 234, figs 8-9. Note punctuation in the above example: comma separates author’s name and year “semicolon separates more than one reference by the same author full stop separates references by different authors figures of plates are enclosed in parentheses to distinguish them from text-figures dash, not comma, separates consecutive numbers Synonymy arrangement according to chronology of bibliographic references, whereby the year is placed in front of each entry, and the synonym repeated in full for each entry, is not acceptable. In describing new species, one specimen must be designated as the holotype; other speci- mens mentioned in the original description are to be designated paratypes; additional material not regarded as paratypes should be listed separately. The complete data (registration number, depository, description of specimen, locality, collector, date) of the holotype and paratypes must be recorded, e.g.: Holotype SAM-—A13535 in the South African Museum, Cape Town. Adult female from mid- tide region, King’s Beach Port Elizabeth (33°51’S 25°39’E), collected by A. Smith, 15 January 1973. Note standard form of writing South African Museum registration numbers and date. 7. SPECIAL HOUSE RULES Capital initial letters (a) The Figures, Maps and Tables of the paper when referred to in the text 6 e.g. *... the Figure depicting C. namacolus ...’; *. ..in C. namacolus (Fig. 10)... (b) The prefixes of prefixed surnames in all languages, when used in the text, if not preceded by initials or full names e.g. Du Toit but A.L.du Toit; Von Huene but F. von Huene (c) Scientific names, but not their vernacular derivatives e.g. Therocephalia, but therocephalian Punctuation should be loose, omitting all not strictly necessary Reference to the author should be expressed in the third person Roman numerals should be converted to arabic, except when forming part of the title of a book or article, such as ‘Revision of the Crustacea. Part VIII. The Amphipoda.’ Specific name must not stand alone, but be preceded by the generic name or its abbreviation to initial capital letter, provided the same generic name is used consecutively. Name of new genus or species is not to be included in the title: it should be included in the abstract, counter to Recommendation 23 of the Code, to meet the requirements of Biological Abstracts. F. E. GRINE RELATIVE SIZES OF THE MAXILLARY DECIDUOUS CANINE AND CENTRAL INCISOR TEETH IN THE KALAHARI SAN (BUSHMAN) AND SOUTH AFRICAN NEGRO MAY 1982 | ISSN 0303-2515 On NAW JUL 4 4 1982 LIBRARIES 4 OF THE SOUTH AFRICAN - MUSEUM | CAPE TOWN INSTRUCTIONS TO AUTHORS 1. MATERIAL should be original and not published elsewhere, in whole or in part. 2. LAYOUT should be as follows: (a) Centred masthead to consist of Title: informative but concise, without abbreviations and not including the names of new genera or species Author’s(s’) name(s) Address(es) of author(s) (institution where work was carried out) Number of illustrations (figures, enumerated maps and tables, in this order) (b) Abstract of not more than 200 words, intelligible to the reader without reference to the text (c) Table of contents giving hierarchy of headings and subheadings (d) Introduction (e) Subject-matter of the paper, divided into sections to correspond with those given in table of contents (f) Summary, if paper is lengthy (g) Acknowledgements (h) References Gi) Abbreviations, where these are numerous 3. MANUSCRIPT, to be submitted in triplicate, should be typewritten and neat, double spaced with 2,5 cm margins all round. First lines of paragraphs should be indented. Tables and a list of legends for illustrations should be typed separately, their positions indicated in the text. All pages should be numbered consecutively. Major headings of the paper are centred capitals; first subheadings are shouldered small capitals; second subheadings are shouldered italics; third subheadings are indented, shouldered italics. Further subdivisions should be avoided, as also enumeration (never roman numerals) of headings and abbreviations. Footnotes should be avoided unless they are short and essential. Only generic and specific names should be underlined to indicate italics; all other marking up should be left to editor and publisher. 4. ILLUSTRATIONS should be reducible to a size not exceeding 12 « 18 cm (19 cm including legend); the reduction or enlargement required should be indicated; originals larger than 35 x 47 cm should not be submitted; photographs should be rectangular in shape and final size. A metric scale should appear with all illustrations, otherwise magnification or reduction should be given in the legend; if the latter, then the final reduction or enlargement should be taken into consideration. All illustrations, whether line drawings or photographs, should be termed figures (plates are not printed; half-tones will appear in their proper place in the text) and numbered in a single series. Items of composite figures should be designated by capital letters; lettering of figures is not set in type and should be in lower-case letters. The number of the figure should be lightly marked in pencil on the back of each illustration. 5. REFERENCES cited in text and synonymies should all be included in the list at the end of the paper, using the Harvard System (ibid., idem, loc. cit., op. cit. are not acceptable): (a) Author’s name and year of publication given in text, e.g.: ‘Smith (1969) describes...’ ‘Smith (1969: 36, fig. 16) describes...’ ‘As described (Smith 1969a, 1969b; Jones 1971)’ ‘As described (Haughton & Broom 1927)...’ ‘As described (Haughton et al. 1927)...’ Note: no comma separating name and year Dagination indicated by colon, not p. names of joint authors connected by ampersand et al. in text for more than two joint authors, but names of all authors given in list of references. (b) Full references at the end of the paper, arranged alphabetically by names, chronologically within each name, with suffixes a, b, etc. to the year for more than one paper by the same author in that year, e.g. Smith (1969a, 19695) and not Smith (1969, 1969a). For books give title in italics, edition, volume number, place of publication, publisher. For journal article give title of article, title of journal in italics (abbreviated according to the Worid list o, scientific periodicals. 4th ed. London: Butterworths, 1963), series in parentheses, volume number, part number (only if independently paged) in parentheses, pagination (first and last pages of article). Examples (note capitalization and punctuation) BULLOUGH, W. S. 1960. Practical invertebrate anatomy. 2nd ed. London: Macmillan. FISCHER, P.—H. 1948. Données sur la résistance et de le vitalité des mollusques. J. Conch., Paris 88: 100-140. FiscHER, P.-H., DuvAL, M. & Rarry, A. 1933. Etudes sur les échanges respiratoires des littorines. Archs Zool. exp. gén. 74: 627-634. \ Konn, A. J. 1960a. Ecological notes on Conus (Mollusca: Gastropoda) in the Trincomalee region of Ceylon. Ann, Mag. nat. Hist. (13) 2: 309-320. Konn, A. J. 19606. Spawning behaviour, egg masses and larval development in Conus from the Indian Ocean. Bull. Bingham oceanogr. Coll. 17 (4): 1-51. THIELE, J. 1910. Mollusca: B. Polyplacophora, Gastropoda marina, Bivalvia. In: SCHULTZE, L. Zoologische und anthropologische Ergebnisse einer Forschungsreise im westlichen und zentralen Siid-Afrika 4: 269-270. Jena: Fischer. Denkschr. med.-naturw. Ges. Jena 16: 269-270. (continued inside back cover) ANNALS OF THE SOUTH AFRICAN MUSEUM ANNALE VAN DIE SUID-AFRIKAANSE MUSEUM Volume 86 Band May 1982 Mei Part 8 Deel THE OSTEOLOGY AND RELATIONSHIPS OF TANGASAURUS MENNELLI HAUGHTON (REPTILIA, EOSUCHIA) By PHILIP J. CURRIE Cape Town Kaapstad The ANNALS OF THE SOUTH AFRICAN MUSEUM are issued in parts at irregular intervals as material becomes available Obtainable from the South African Museum, P.O. Box 61, Cape Town 8000 Die ANNALE VAN DIE SUID-AFRIKAANSE MUSEUM word uitgegee in dele op ongereelde tye na gelang van die beskikbaarheid van stof Verkrygbaar van die Suid-Afrikaanse Museum, Posbus 61, Kaapstad 8000 OUT OF PRINT/UIT DRUK 12 (1= 3s 5-8) 31-2 4-5 8) tp) Ser Op 6(l, tpi), 70=4), 8: 9-2, 7), 10-3); 141-2, 5, 7, t=p.i.), 15(4-5), 24), 27, 314=3), 32), 33, 45) EDITOR/REDAKTRISE Ione Rudner Copyright enquiries to the South African Museum Kopieregnavrae aan die Suid-Afrikaanse Museum ISBN 0 86813 029 X Printed in South Africa by In Suid-Afrika gedruk deur The Rustica Press, Pty., Ltd., Die Rustica-pers, Edms., Bpk., Court Road, Wynberg, Cape Courtweg, Wynberg, Kaap THE OSTEOLOGY AND RELATIONSHIPS OF TANGASAURUS MENNELLI HAUGHTON (REPTILIA, EOSUCHIA) By PHILIP J. CURRIE Provincial Museum of Alberta, Edmonton (With 6 figures and 2 tables) [MS accepted 11 November 1981] ABSTRACT The osteology of Tangasaurus mennelli, an eosuchian reptile from the Upper Permian of Tanzania, reveals a number of adaptations for an aquatic existence. Specimens from Madagas- car that were attributed to Tangasaurus mennelli represent a distinct, as yet unnamed, genus that is related to Tangasaurus but less specialized. Tangasaurus is closely related to, but less specialized for an aquatic life than Hovasaurus boulei of the Upper Permian of Madagascar. The taxonomy of younginids and tangasaurids is re-examined, and it is concluded that these families together comprise a distinct taxonomic unit, the Younginoidea, that did not give rise to any other eosuchians. CONTENTS PAGE UTD ANE DLO MMR arte eet alates ay este. Aa: Fee eee res, heirs or oh Meg ode de MEN en eae 247 BBY SUSHMIO UOMO ese chs his SE aioe e eet vance a 5 Se GAC Maange, coats ae eee Ue 252 SK UMPIRE ey cir lini», coane Kets aire iy hw Gienarls se uietane “atin avai Maus ee Sotayitn Mea Ne eas 252 Vere Nl CRIN GL TOS Siena tes Ad. ial op wid ONE ato saa ated aha te URaaee ee lee ous Ce aa 253 PMP MEMMICU ATESKClECOMY § cesar: eeinic a an tees Risa eo ene cne be eee 256 PN OIA ONT MAPMT e ete. e csc e sean lee rua ce ieee conven ena s auge es cnn See eae 260 Supeniamily ounginoidea superfam. NOV. .....5...2..4..5-42--+- 284504: 260 NSTI MVCOUMP TTI AEM crysis aa ess ap geeve eda sk steals te santas leternmgetane, neatenclaw weer: 260 HAMMAM ASAT ACs rai okies, Seis ee ulm tacos ge Mage ate area eo 262 Subtamily;Kenyasaurinae subfam\: MOV. .45>...5.:2-c40es.- 45-52 ose. +=: 262 Sulbbtanimlyeltan eas aUnnacr.. os caac mlat os ccace eeoettge oa ae oe Reedley aye as 262 (COME SIONS sg ache leo eo atk eee Caen rel ea aR ees tn a) nce eer ie ce AE 263 PMC OLE CCIM CIIES Hcl ial nts tee sia eaeeelern ty bude mune sentae habeas slays Mech ateeaee aides 264 ANCHE MN COSUM PECs eMac eeuiies meaty sldye eae vaca Geet eae Ace cy nce Ata tie Ye 264 INTRODUCTION Eosuchians are generally considered to be the basal stock from which all later diapsids evolved. Recent papers by Carroll (1975, 1976a, 1976b, 1977, 1978), Currie (1980, 1981), Gow (1975), Harris & Carroll (1977), and Reisz (1977) have shown that eosuchians originated earlier and were more diverse than previously suspected. By the end of the Permian, at least three distinct lines had developed. One line is characterized by cervical elongation as in Prolacerta (Gow 1975) and the highly specialized Tanystropheus (Wild 1973). A separate lineage of eosuchians might have given rise to lizards (Carroll 1975) and the 247 Ann. S. Afr. Mus. 86 (8), 1982: 247-265, 6 figs, 2 tables. 248 ANNALS OF THE SOUTH AFRICAN MUSEUM gliding genus Daedalosaurus (Carroll 1978). Youngina is one of a diverse assemblage of terrestrial and aquatic forms that appears to have given rise to archosaurs (Carroll 1976a) and sauropterygians (Carroll 1981). The more de- tailed relationships of some of these taxa are considered here. Four genera of reptiles seem to share a common ancestry with Youngina: Tangasaurus (Haughton 1924), a Late Permian reptile specialized for an aquatic existence, known from Tanzania; Kenyasaurus (Harris & Carroll 1977), a terrestrial relative of Tangasaurus from Lower Triassic strata of Kenya; and ‘Datheosaurus’* and Hovasaurus from Upper Permian beds of Madagascar. ‘Datheosaurus’ was a terrestrial form like Kenyasaurus, whereas Hovasaurus was even more specialized for an aquatic existence than was Tangasaurus. Descrip- tions have recently been published on Youngina (Gow 1975) and Kenyasaurus (Harris & Carroll 1977), and papers on ‘Datheosaurus’ and Hovasaurus are being prepared by Carroll and Currie respectively. The purpose of this paper is to redescribe the known specimens of Tangasaurus and to compare them with other tangasaurids. In 1924 S. H. Haughton described two fossil reptile specimens that had been collected from Upper Permian strata in the vicinity of Tanga, in what is now Tanzania. These were recognized as a previously unknown taxon and named Tangasaurus mennelli. Although the specimens were designated as co-types, the smaller, better preserved one (Fig. 1A) is here considered as the lectotype. It is in the museum in Bulawayo, Zimbabwe, while its counterpart, SAM-—6231 (Fig. 1B), and the larger specimen, SAM-6232 (Fig. 1C), are in the South African Museum, Cape Town. On the basis of postcranial characteristics, Haughton (1924) felt that Tangasaurus was probably a diapsid reptile that, because of the long, powerful, flattened tail, had become adapted for an aquatic existence. Numerous specimens of small reptiles had been collected in beds of approxi- mately the same age along the Sakamena River of southern Madagascar. One of the animals collected was described as Broomia madagascariensis (Piveteau 1925), but was referred the following year by Piveteau to Tangasaurus menelli (sic). Contrary to Haughton, Piveteau felt that Tangasaurus was a primarily terrestrial animal. Following Nopcsa (1924), he considered it to be related to Araeoscelis, Kadaliosaurus, Broomia, Saurosternon, and Pleurosaurus. A new genus and species, Hovasaurus boulei, also from Madagascar, described by Piveteau (1926) was considered to be related to Mesosaurus. Although not as specialized as Mesosaurus for living in the water, Piveteau felt that the short neck, short manus, well developed haemal spines and slight pachyostosis of the ribs showed that Hovasaurus was an aquatic animal. Haughton (1930) restudied Piveteau’s specimens from Madagascar, con- cluding that Tangasaurus and Hovasaurus were allied and that both were diapsids. Hovasaurus was recognized as a reptile adapted to an aquatic existence through reduction of the forelimb and coracoid, and retarded ossification and * While this paper was in press ‘Datheosaurus’ specimens were redescribed as Thadeosaurus (Carroll 1981). 249 TANGASAURUS MENNELLI HAUGHTON "WD | = d[BIG “MOIA [RIUDA “ZEZI-NIVS “OD ‘(LLOT MOULD 2 Sle] Wo.) uowuoods Surposoid yo yaediojunos ‘TEZI-WVS “A ‘OMQRquIZ ‘wnosnpy OARML[NG Ut 9d{10}99| JO MOIA [RIIUDA “YW “Yjauuaw snanvspsuvy, “| “Bt q WV 250 ANNALS OF THE SOUTH AFRICAN MUSEUM elongation of the body. Tangasaurus was considered to be morphologically intermediate between Youngina and Hovasaurus. Other specimens described by Piveteau (1926) were not associated with Tangasaurus and Hovasaurus until recently. A large number of caudal sections were recovered along the Sakamena River of Madagascar. The caudal vertebrae have high neural spines, and the haemal spines are long and plate-like. Piveteau recognized these as belonging to a reptile adapted to swimming, but did not have any other skeletal elements associated with the caudal vertebrae. These are now known to belong to Hovasaurus. A third genus recognized by Piveteau (1926: 171-172) was assigned with doubt to the European genus Datheosaurus, now considered to be congeneric with Haptodus (Currie 1979), a sphenacodont pelycosaur. The specimens refer- red to this genus were restudied recently by Carroll (1981) who considers them to be a new genus. Morphologically this animal appears to be close to the ancestral stock of Tangasaurus and Hovasaurus. Kenyasaurus mariakeniensis from the Lower Triassic of Kenya (Harris & Carroll 1977) is considered to be most closely comparable with Tangasaurus and Hovasaurus on the basis of general body form, the presence of a sternum and, particularly, the anatomy of the foot. Piveteau (1926) included Broomia, Saurosternon and Tangasaurus in the Tangasaurinae (sic). After Haughton’s paper of 1930 demonstrated the anatom- ical similarities between Tangasaurus and Hovasaurus, these genera were usually included as the only representatives of the family Tangasauridae (Camp 1945; Romer 1956, 1966; Piveteau 1955; Orlov 1964; Kuhn 1969). Depending on how the various authors have classified primitive diapsid reptiles, the Tangasauridae have been referred to the Araeosceloidea (Nopcsa 1924; Piveteau 1926), Eosuchia (Haughton 1930; Von Huene 1940, 1952; Piveteau 1955; Romer 1956, 1966; Kuhn 1969), Prolacertilia (Watson 1957; Orlov 1964) or Protorosauria (Camp 1945). In re-examining the anatomy of tangasaurids, it became obvious that there is a great deal of confusion concerning the identification and anatomy of these animals. The anatomy of Tangasaurus and Hovasaurus is very similar, and there are few characters to distinguish these genera. Although more than 300 tanga- saurid specimens were collected in Madagascar, in almost every case two of the most diagnostic parts of the body—the skull and the tail—were missing. The problem is compounded by poor preservation of the type specimens of Tanga- saurus mennelli from Tanzania that generally have been ignored in favour of the better preserved specimens from Madagascar. Finally, many of the specimens have been misidentified because they are immature and show few distinctive characters. As part of a revision of tangasaurid anatomy, the majority of specimens that have been figured were re-examined and, in many cases, re- identified (Table 1). Hopefully, the confusion concerning the identification of tangasaurids has been resolved. This will permit more accurate conclusions concerning evolutionary and developmental lineages, palaeoecology, geographic Identification (this paper) ‘Datheosaurus’ ‘Datheosaurus’ ‘Datheosaurus’ ‘Datheosaurus’ ‘Datheosaurus’ ‘Datheosaurus’ Hovasaurus boulei H. boulei . boulei . boulei . boulei . boulei boulei boulei boulei boulei boulei boulei boulei boulei copes FT Te TF PTT . boulei H. boulei H. boulei H. boulei ?H. boulei Kenyasaurus mariakaniensis Tangasaurus mennelli T. mennelli Not a tangasaurid Not a tangasaurid Not a tangasaurid Not a tangasaurid TANGASAURUS MENNELLI HAUGHTON Pa TABLE 1 Figured specimens of Tangasauridae. Specimen number MNHN 1908-5-1 MNHN 1908-11-4 MNHN 1908-11-5 MNHN 1908-11-6 MNHN 1908-11-7 MNHN 1908-11-19 MNHN 1908-21-2 MNHN 1908-21-7 MNHN 1908-21-8 MNHN 1908-21-10 MNHN 1908-21-11 MNHN 1908-21-16 MNHN 1908-32-22 MNHN 1908-32-23 MNHN 1908-32-24 MNHN 1908-32-25 MNHN 1908-32-26 MNHN 1908-32-29 MNHN 1925-5-30 MNHN 1925-5-31 MNHN 1925-5-32 MNHN 1925-5-33 MNHN 1925-5-34 MNHN, number unknown MNHN, number unknown KNM-MA1 SAM-6231 SAM-6232 MNHN 1909-3-—30 MNHN, number unknown MNHN, number unknown MNHN, number unknown Previous identification Tangasaurus mennelli (Piveteau 1926, pl. 12 (fig. 1)) T. mennelli (Piveteau 1926, pl. 10 (fig. 2)) T. mennelli (Piveteau 1926, pl. 11 (fig. 2), text-figs 17-18; Piveteau 1955, fig. 9; Gladstone & Wakeley 1932 fig. 2; Kuhn 1969, figs 18-6, 18-8) T. mennelli (Piveteau 1926, pl. 16 (fig. 2)) T. mennelli (Piveteau 1926, pl. 10 (fig. 1)) ? Datheosaurus sp. (Piveteau 1926, pl. 17 (fig. 2)) Hovasaurus boulei (Piveteau 1926, pl. 8 (fig. 2)) H. boulei (Piveteau 1926, pl. 8 (fig. 1); Haughton 1930, fig. 3C) H. boulei (Piveteau 1926, pl. 7 (fig. 1)) T. mennelli (Harris & Carroll 1977, text-fig. 5B) Indeterminate reptile (Piveteau 1926, pl. 10 (fig. 4)) T. mennelli (Piveteau 1926, pl. 15 (fig. 5) text-fig. 20; Gladstone & Wakeley 19372, text-fig. 2) H. boulei (Piveteau 1926, pl. 9 (fig. 4), text-fig. 14) H. boulei (Piveteau 1926, pl. 9 (figs 1-2, text-fig. 16) T. mennelli (Piveteau 1926, pl. 13 (fig. 1); Harris & Carroll 1977, text-fig. 5A) T. mennelli (Piveteau 1926, pl. 14 (fig. 3)) T. mennelli (Piveteau 1926, pl. 15 (fig. 1)) H. boulei (Piveteau 1926, pl. 7 (fig. 2)) H. boulei (Piveteau 1926, pl. 9 (fig. 3), text-fig. 12) T. mennelli (Piveteau 1926, pl. 15 (fig. 4), text-fig. 22; Piveteau 1955, text-fig. 10) T. mennelli (Piveteau 1926, pl. 16 (fig. 1), text-fig. 23; Haughton 1930, text-fig. 4C; Piveteau 1955, text-fig. 11; Kuhn 1969, text-fig. 18-7; Harris & Carroll 1977, text-fig. 5C) T. mennelli (Piveteau 1926, pl. 10 (fig. 3)) T. mennelli (Piveteau 1926, pl. 14 (fig. 2); Camp 1945, text-fig. 10) H. boulei (Piveteau 1926, pl. 7 (fig. 3); Haughton 1930, text-fig. 1A) T. mennelli (Piveteau 1926, pl. 13 (fig. 5), pl. 14, fig. 1) Kenyasaurus mariakaniensis (Harris & Carroll 1977, text-figs 1-4) Tangasaurus mennelli (Haughton 1924, pl. 2, text- figs 1-3; Von Huene 1926, text-fig. 33; Romer 1956, text-fig. 186E; Orlov 1964, text-fig. 468; Kuhn 1969, text-figs 17-12; Harris & Carroll 1977, text-fig. 6) . mennelli (Haughton 1924, pl. 1) . mennelli (Piveteau 1926, pl. 15 (figs 2-3)) . mennelli (Piveteau 1926, pl. 13 (figs 2, 2A)) . mennelli (Piveteau 1926, pl. 13 (fig. 3)) Se oe es . mennelli (Piveteau 1926, pl. 13 (figs 4, 4A)) US ANNALS OF THE SOUTH AFRICAN MUSEUM distribution and stratigraphy. One unfortunate result is that Tangasaurus can no longer be used for biostratigraphic correlation of the middle division of the Tanga beds of Tanzania with the Lower Sakamena Formation of Madagascar (McKinlay 1956, 1960). Tangasaurid eosuchians were relatively small lizard-like reptiles. The largest is Hovasaurus with an estimated maximum snout—vent length of about 30 cm. The largest specimen of Tangasaurus is 20 per cent smaller than the largest Hovasaurus, but it can be inferred from the incompleteness of ossification that larger specimens probably existed. The linear dimension of a mature specimen of ‘Datheosaurus’ are about 35 per cent smaller than the same dimensions in Hovasaurus. Proportions and lengths listed in this paper for Hovasaurus and ‘Datheosaurus’ were estimated for adult size from the allometric growth curves. Hovasaurus and Kenyasaurus are valid genera. Use of the name Tanga- saurus in this paper is restricted to the two specimens from Tanzania. When referring to generic characters attributed by Piveteau (1926), Haughton (1930), and others to specimens from both Tanzania and Madagascar, the name Janga- saurus will be used in quotation. ‘Datheosaurus’ is used to refer to most of the specimens from Madagascar that Piveteau (1926) identified as Datheosaurus and Tangasaurus. The following abbreviations have been used when giving specimen num- bers: MNHN, Muséum National d’Histoire Naturelle SAM, South African Museum DESCRIPTION SKULL Cranial anatomy is poorly known in tangasaurids. In one of the two known specimens of Tangasaurus, the poorly preserved skull is exposed in palatal aspect. The bone is crushed flat, and seems little better than a film of carbon. Further preparation of the specimen seems unfeasible. The anterior portion of the skull was not collected, but the preserved portion is 38 mm in length. Haughton (1924) felt that the full length of the skull would have been between 50 and 60 mm. In the light of the anterior tapering of the skull, there was probably no elongate snout, and it is doubtful that the skull would have exceeded 45 mm. The only bone that can be identified with any degree of certainty is the basisphenoid-parasphenoid complex. The cultriform process is long and tapering, and the basipterygoid processes appear to be short. The complex is concave ventrally in transverse section between the tuberosities. Haughton (1924) made observations on the pterygoid and various palatal vacuities, but none of these can be seen clearly enough in the specimen to merit description. The basioccipital can be seen at the back of the skull, but shows no distinctive characters. Von Huene (1926) identified one bone as either stapes or TANGASAURUS MENNELLI HAUGHTON 253 quadrate, but whatever it is, the bone gives us no better understanding of the skull. Only the postorbital and palatal regions of the skull of Hovasaurus are known. The contours and proportions of the parasphenoid-basisphenoid com- plex of Tangasaurus are similar to those of Hovasaurus. VERTEBRAE AND RIBS Haughton (1924) estimated that there were eighteen presacral vertebrae in the lectotype (SAM-6231) of Tangasaurus mennelli. This number is much less than that of other eosuchians, and it is assumed that he had not included the cervical vertebrae in his total. Because he stated that there were seven or fewer anterior vertebrae obscured by bones of the pectoral girdle, his estimated presacral count can be raised to a maximum of twenty-five. This is the number of vertebrae found in Hovasaurus. As in Hovasaurus, the neck of Tangasaurus is short—probably including only five vertebrae. Primitive diapsid reptiles characteristically have two sacral vertebrae. Haughton (1924: 3) stated that one specimen of Tangasaurus mennelli (SAM- 6231) had three fused sacral vertebrae. However, preservation in the sacral region of this specimen is poor, and it is likely that he misinterpreted the first caudal rib as a sacral rib. In a later paper Haughton (1930) stated that ‘Tangasaurus’ has two sacrals. The total length of the tail is known in few eosuchians. 24 caudal vertebrae are preserved in the lectotype of Tangasaurus, and 28 in SAM-6232. However, the tail was clearly much longer than this. ‘Datheosaurus’ has at least 45 caudals and Hovasaurus probably had more than 70. Details of the dorsal vertebrae are difficult to delineate in the type speci- mens of Tangasaurus. In Hovasaurus, there is a process at the base of the neural spine that acts as an extra intervertebral articulation (Fig. 2). A similar accessory process appears to be present in at least one of the dorsal vertebrae of the larger specimen of Tangasaurus. Contact between the neural spines has also been noted in Youngina (Currie 1981), ‘Datheosaurus’, and Kenyasaurus. Specimens of Tangasaurus (Figs 3A—-B, 4) have caudal vertebrae that are specialized for propulsion in water. The neural spines are high and the haemal spines are long and expanded distally. The neural spines of the mid-caudals are higher than the presacral neural spines, but are not as long as the haemal spines. The specialization is not as great as in Hovasaurus (Fig. 3D) in that the neural spines are relatively lower, only about 35 per cent greater than the length of the associated centrum compared with more than 125 per cent in Hovasaurus. Nevertheless, the caudal specialization suggests that Tangasaurus and Hova- saurus are Closely related. Specimens from Madagascar attributed to ‘Tanga- saurus’ and ‘Datheosaurus’ have unspecialized caudal vertebrae with low neural spines (Fig. 3C). The dorsal ribs of Tangasaurus have a single head and are not pachyostotic. 254 ANNALS OF THE SOUTH AFRICAN MUSEUM Fig. 2. Hovasaurus boulei, MNHN 1908-32-59. Abbreviations: ac—accessory articulation on neural spine, b—ballast, COR—coracoid, e—entepicondyle, g—gastralia, H—humerus, i—intermedium, lc—lateral centrale, m—mammillary process, mc—medical centrale, pi—pisiform, R—radius, r—radiale, ST—sternum, U—ulna, u—ulnare. Scale = 1 cm. TANGASAURUS MENNELLI HAUGHTON 255 Fig. 3. Caudal vertebrae of tangasaurid eosuchians. A. Tangasaurus (Bulawayo Museum), 19th to 21st, right aspect. B. Tangasaurus, SAM-6232, 20th to 23rd, left view. C. ‘Datheosaurus’, MNHN 1908-5—-1, mid-caudal, left view. D. Hovasaurus, MNHN 1908-32-64, 14th caudal, left aspect. Each scale = 1 cm. 256 ANNALS OF THE SOUTH AFRICAN MUSEUM Fig. 4. Tangasaurus mennelli, SAM-6233, caudal vertebrae. Scale = 1 cm. In Hovasaurus, moderate pachyostosis can be seen in the ribs of large indi- viduals but it is usually not apparent in juveniles. An exact count of caudal ribs is not possible because of the poor preserva- tion of the Tangasaurus specimens. The total number, however, would have been between 9 and 12. There are a maximum of 12 pairs of caudal ribs in Hovasaurus and an additional 2 pairs of transverse processes. Nineteen pairs of caudal ribs and transverse processes are found in specimens from Madagascar that were formerly attributed to Tangasaurus, and 28 pairs in Kenyasaurus. APPENDICULAR SKELETON Both Piveteau (1926) and Haughton (1924, 1930) recognized that the relative lengths of elements of the appendicular skeleton change with age. More precise calculations (Currie in preparation) have shown that ‘Tangasaurus’ specimens of Madagascar are juveniles of ‘Datheosaurus’, and that the allometric growth coefficients for ‘Datheosaurus’ and Hovasaurus are different (Fig. 5). The two specimens of Tangasaurus represent animals of different ages and consequently show some differences in limb proportions. Comparison can be made with Hovasaurus by means of the allometric growth equations. The constants b’ and k,,’ power equation y = b’xtyx (Currie 1978) were solved to describe a growth series of Hovasaurus boulei. The average length of a dorsal centrum that serves as the base for comparison is represented by x, while the length of the element being compared is represented by y. From this informa-. tion, the expected mean length (in millimetres) of each element was computed for Hovasaurus specimens of the same size as the Tangasaurus specimens SAM-— 6231 (x = 6,6 mm) and SAM-6232 (x = 8,0 mm). Lengths of the metacarpals and metatarsals were not included in the lengths of the digits of the manus and TANGASAURUS MENNELLI HAUGHTON DF 1,50 0,50 1,00 Fig. 5. Relationship between humerus length (ordinate) and length of associated thoracic centrum (abscissa) in ‘Datheosaurus’ (circles) and Hovasaurus (squares). Measurements con- verted to logarithms and plotted on arithmetic paper. Differences in slope represent differences in allometric growth coefficients. 258 ANNALS OF THE SOUTH AFRICAN MUSEUM pes. With the exception of the length of metatarsal IV of SAM-—6231 and the length of digit IV of the pes of SAM-—6232, all measurements fell within the 95 per cent confidence intervals for these dimensions in Hovasaurus (Table 2). This shows that the relative lengths of limb elements are almost the same in Tangasaurus and Hovasaurus, which also suggests close relationship because of the specialized proportions of Hovasaurus limbs for swimming. In the smaller specimen of Tangasaurus (SAM-6231), the humerus is shorter than the femur, whereas the femur is the shorter of the two in the larger specimen. This suggests that the humerus grew faster than the femur as the animal increased in size. The same thing happens in both ‘Datheosaurus’ and Hovasaurus where more complete growth series are known. Haughton (1930) stated that the limbs of ‘Tangasaurus’ are longer relative to the body than those of Hovasaurus. However, it is now known that the humerus of Tangasaurus is 5,8 times the length of a dorsal centrum (x), that of a mature ‘Datheosaurus’ is 6,7x, and that of a mature specimen of Hovasaurus is 7,1x. Similarly, other limb elements of Tangasaurus have slightly smaller relative lengths than the same elements in Hovasaurus. In Tangasaurus, the radius is 60 per cent the length of the humerus, TABLE 2 Comparisons of the lengths of postcranial elements of Hovasaurus boulei and Tangasaurus mennelli. The constants b’ and k,,’ of the power equation y = b’x‘¥x’ have been solved to describe growth series of H. boulei. From this information, the expected mean lengths (in mm) of each element have been computed for H. boulei specimens the same size as the co-types of T. mennelli (x = 6,6 mm for SAM-6231, 8,0 mm for SAM-6232). Lengths of the metacarpals and metatarsals are not included in the lengths of the digits of the manus and pes. Abbrevia- tions: N—size of sample, R—correlation coefficient. Estimated mean value of y | Measured for H. boulei value of y = length of N R ray b’ X y 95% confidence y for T. : interval mennelli Corac0id: %.,i63 2: 9 0,981 2,229 0,360 6,6 24,1 20,7-28,0 2553 8.0 37,0 30,8-44,6 Si E7/ SLCRMUIM 682.ae arg 15 0,987 1,780 0,739 6,6 22.1 20,7-23,7 21,8 8,0 29,9 27,4-32,7 30,3 IAUMETUS Ga ete 20 0,993 1,638 1,674 6,6 36,8 35,6-38,1 36,0 8,0 50,5 48,4-52,7 48,5 Radiusevat ere 13 0,994 1,306 1,893 6,6 22:2. 21,2-23,3 21,6 8,0 28,6 26,9-30,3 28,0 MetacarpalIV.... 12 0,991 1,354 0,599 6,6 dest 7,2- 8,2 8,0 Digit IV (manus) . 12 0,983 1,236 2,035 6,6 20,9 19,2-22,7 21,0 hme eee eee 14 0,996 1,067 2,952 6,6 2251 21,1-23,1 23,0 RemMUIe ee 16 0,990 1,334 3,146 6,6 39,0 36,9-41,2 39,0 8,0 50,4 47,1-54,0 47,2 RIDA haere 15 0,991 1,298 2,872 6,6 33%) 31,3-35,1 34,0 8,0 42,6 39,6-46,0 40,0 MetatarsalIV .... 14 0,995 1,340 1,310 6,6 16,4 15,6-17,2 17,4 8,0 DED 20,0-22,6 20,5 Digit IV (Pes) .... 9 0,991 1,493 1,996 6,6 33,3 30,8-36,2 32,2 8,0 44,5 399-495 39,1 TANGASAURUS MENNELLI HAUGHTON 259 whereas in mature Hovasaurus it is only 52 per cent. In the relatively unspecial- ized ‘Datheosaurus’ the radius is 54 per cent the length of the humerus at maturity. Relative to the length of a dorsal centrum, however, the length of the forearm of Tangasaurus is only 3,3x, whereas that of Hovasaurus is the same as ‘Datheosaurus’ (3,7x). This contradicts Haughton’s (1930) statement that the forearm is relatively shorter in Hovasaurus. It should be remembered that his statement is true for immature specimens, and he did not have a complete growth series available to him. The tibia is 85 per cent the length of the femur in Tangasaurus and Hovasaurus at maturity, and 90 per cent in ‘Datheosaurus’. Many eosuchians, including Youngina (Broom 1922), Tangasaurus (Fig. 1), Hovasaurus (Piveteau 1926), ‘Datheosaurus’ (Piveteau 1926) and Kenyasaurus (Harris & Carroll 1977), have ossified sterna. The dimensions and outline of the sternum of Tangasaurus fall within the range of Hovasaurus. The coracoids of Tangasaurus and Hovasaurus are similar in outline. Haughton (1930) stated that they are distinguishable on the basis of proportions, but the coracoids of the type specimen of Tangasaurus mennelli are too poorly preserved to confirm this statement. The humeri of Tangasaurus are identical in outline to those of many specimens of Hovasaurus. Well-ossified humeri of Tangasaurus, Hovasaurus, and ‘Datheosaurus’ share the distinctive feature of a greatly expanded entepicon- dyle (Figs. 1-2). The width of the distal end of the humerus is up to 40 per cent of its length. It is worth noting that the known humeri of Youngina are all from immature animals and consequently do not show a well-ossified entepicondyle. However, the curvature at the base of the entepicondyle and the position of the entepicondylar foramen suggest that the humerus of mature individuals of Youngina also had a greatly expanded entepicondyle. The tarsus of Tangasaurus, Hovasaurus, Kenyasaurus, and ‘Datheosaurus’ is specialized in the loss of the fifth distal tarsal as a discrete element. Harris & Carroll (1977) refer to a specimen of Hovasaurus (MNHN 1908-21-10) and suggest that the fifth and fourth distal tarsals fuse at maturity. A very distinctive characteristic of most specimens of Hovasaurus is the presence of abundant pebbles, mainly quartz, in the abdominal cavity (Fig. 2). It was assumed (Haughton 1930) that these were gastroliths, but they seem to be too numerous, large, and closely packed to have functioned effectively in food breakdown. Considering the apparent aquatic habits of Hovasaurus, it may be more appropriate to consider them as having served as ballast. The same function appears to be true for ‘gastroliths’ of crocodiles (Cott 1961) and plesiosaurs (Darby & Ojackangas 1980). Stomach stones are found in at least one specimen of ‘Datheosaurus’ (MNHN 1908-11-5), (Piveteau 1926, pl. 11) but are few in number and possibly did serve as aids to digestion. When present, the relative abundance of stomach stones is a quick way to distinguish Hovasaurus from ‘Datheosaurus’. Absence of stomach stones from the Tangasaurus speci- mens does not necessarily mean that this genus did not swallow pebbles. Some 260 ANNALS OF THE SOUTH AFRICAN MUSEUM specimens of Hovasaurus do not have stones in the abdomen, so this may possibly be explained as post-mortem rupture of the abdominal cavity before the cadaver came to rest and was buried by sediment. TAXONOMY In the light of improved knowledge of the morphology of the tangasaurids, it is possible to establish a diagnosis for the family. The relationship of the Tangasauridae to Youngina will be considered here as well. In recent years, it has become widely accepted that a proper taxonomic diagnosis should emphasize derived (advanced) character states rather than the retention of primitive characters. Characters listed in the following suprageneric diagnoses are derived and can be used to distinguish the taxa from all known eosuchian taxa that are not considered in this paper. Numbers in parentheses refer to Figure 6. Class REPTILIA Linnaeus, 1758 Subclass LEPIDOSAURIA Dumeéril & Bibron, 1839 Order EOSUCHIA Broom, 1914 Suborder YOUNGINIFORMES Romer, 1945 Superfamily YOUNGINOIDEA superfam. nov. Fig. 6 Diagnosis Distinctive sutures on parietal for frontal and postfrontal (1). Aceessory intervertebral articulation present on mid-line of dorsal neural spine (2). Co- ossification of paired sternal plates into a single unit in mature animals (3). Entepicondyle of humerus strongly developed at maturity (4). Olecranon pro- cess and sigmoidal notch of ulna poorly developed in mature animals (5); radius longer than shaft of ulna (6). Family Younginidae Broom, 1914 Youngina Broom, 1914 Diagnosis 3 premaxillary, 15-23 maxillary, and 20 dentary teeth. Zygapophyses of anterior dorsal vertebrae extend laterally beyond the centra and inclined at low angle from horizontal; neural spines low and rectangular. Iliac blade short and almost vertical. Humerus only about 70% length of femur, compared with 75 % in immature Hovasaurus and 110% in mature Hovasaurus. Radius 80% length of humerus, and 60 % length of tibia; tibia 90% length of femur. Proximal head of fifth metatarsal expanded but not hooked. 261 TANGASAURUS MENNELLI HAUGHTON ‘1X9] OY] Ul passnosip s19djoeIeYS 0} JoJoI SIOqUINN “SuBTYyONSOS proursunoX jo sdiysuonejesiojul yuoredde oy} Suimoys yeyo onouesojfyd °9 “B14 ae ered ee ease snineseAuay _sninesoauje .___ seulinesebue| ——— seulineseAusy sepiunesebue, ————————" g. eulbuno, sepiulbunod 262 ANNALS OF THE SOUTH AFRICAN MUSEUM Family Tangasauridae Camp, 1945 Diagnosis Humerus as long as or longer than femur in mature animals (7); radius 50-65 per cent length of humerus and 65-75 per cent length of tibia at maturity. Scapula low in lateral aspect and mainly a ventral element (8); coracoid as large as the scapula (9). Medial centrale contacts fourth distal carpal, thereby prevent- ing contact between the lateral centrale and third distal carpal (10). Fifth distal tarsal not a discrete element (11). Subfamily Kenyasaurinae subfam. nov. Diagnosis 19-28 pairs of caudal ribs and transverse processes present, all of which taper distally (12). Kenyasaurus Harris & Carroll, 1977 Type species Kenyasaurus mariakaniensis Harris & Carroll 1977: 140. Diagnosis Low but anteroposteriorly elongate neural spines in the dorsal region; 56 caudal vertebrae; 28 pairs of caudal ribs and transverse processes. Astragalus almost triangular rather than primitive L-shape; pronounced process on fifth metatarsal for insertion of peroneus brevis. Thadeosaurus colcanapi Carroll, 1981 Broomia madagascariensis Piveteau, 1925: 157. Datheosaurus sp., Piveteau, 1926: pl. 17 (fig. 2). Tangasaurus menelli, Piveteau, 1926: pl. 10 (figs 1-2), pl. 11 (figs 1-2), pl. 12 (fig. 1), pl. 16 (fig. 2). Diagnosis Neural spines tall and rectangular in dorsal region; 47 caudal vertebrae; 19 pairs of caudal ribs and transverse processes. Small numbers of gastroliths present in abdominal cavities of some specimens. Subfamily Tangasaurinae Piveteau, 1926 Diagnosis Neural spines high in dorsal region and higher in proximal and mid-caudal regions (13). 9-12 pairs of caudal ribs (14); anterior caudal ribs expanded distally (15); haemal spines large and platelike (16). Presacral intercentra, with the exception of the first three, do not ossify until animal is mature (17). TANGASAURUS MENNELLI HAUGHTON 263 Tangasaurus Haughton, 1924 Type species Tangasaurus mennelli Haughton, 1924: 3. Tanganasaurus Piveteau, 1925: 155. Tangasaurus menelli Piveteau, 1926: 78. Tangasaurus minelli Peyer, 1937: 115. Diagnosis Neural spines of dorsal vertebrae high and rectangular; height of neural spine of mid-caudal vertebra about 35 per cent greater than length of associated centrum, and about 75 per cent length of associated haemal arch and spine. Hovasaurus Piveteau, 1926 Type species Hovasaurus boulei Piveteau, 1926: 78. Diagnosis Skull lacks tabular. 5 cervical, 20 dorsal, 2 sacral and at least 70 caudal vertebrae. Height of neural spines of mid-dorsal vertebrae at least 75 per cent greater than length of associated centrum, whereas height of a mid-caudal neural spine can be more than 125 per cent greater than length of associated centrum; neural spine of mid-caudal vertebra almost 90 per cent length of associated haemal arch and spine. Mammillary processes on neural spines of anterior dorsals. Ribs slightly pachyostotic in mature animals. High number of stones in abdominal cavity suggesting they were used as ballast. CONCLUSIONS Tangasaurus mennelli is represented only by three specimens from the Tanga region of Tanzania. Specimens from Madagascar that have been attri- buted to this species represent a distinct as-yet-unnamed genus that Piveteau (1926) referred to as ‘Datheosaurus’ and that is being described by Carroll (1981). Four genera from Africa and Madagascar are herein assigned to the Tangasauridae. Two subfamilies are recognized on the basis of differences in caudal anatomy. Kenyasaurines were not as highly specialized for an aquatic existence as were the tangasaurines and they were intermediate in morphological specialization between Youngina and Hovasaurus. Tangasaurids and younginids share a number of derived characters and, therefore, have been united into a single superfamily, the Younginoidea. Acerosodontosaurus (Currie 1980), Galesphyrus (Carroll 1976a), and Heleosaurus (Carroll 1976a) have recently been referred to the Younginidae, but they do not possess the derived characters shared by Youngina and the tanga- saurids. Therefore, they should not be considered as younginoids. Their sys- tematic position will be considered in a separate paper. 264 ANNALS OF THE SOUTH AFRICAN MUSEUM The shared derived characters of the Younginoidea are not found as a suite in other eosuchians, which strongly suggests that Youngina is not the ancestral morphotype for any eosuchians other than the Tangasauridae. ACKNOWLEDGEMENTS Casts and photographs of the type specimens of Tangasaurus mennelli were supplied by Dr M. A. Raath (as Executive Director of the National Museums and Monuments of Rhodesia (now Zimbabwe) in 1978) and Dr M. A. Cluver (South African Museum). I am grateful to Dr R. L. Carroll for preparing and casting tangasaurid specimens in the collections of the Muséum National d’His- toire Naturelle and the South African Museum, and for providing me with Figure 1B—C and unpublished data on his research. The manuscript was read by Dr R. L. Carroll, McGill University, Dr M. J. Heaton, University of Toronto, Dr J. A. Hopson, University of Chicago, and Dr R. Reisz, University of Toronto, who made many useful suggestions. Mr Clive Booth (South African Museum) prepared Figure 4. REFERENCES Broom, R. 1914. A new thecodont reptile. Proc. zool. Soc. Lond. 1914: 1072-1077. Broom, R. 1922. An imperfect skeleton of Youngina capensis, Broom, in the collection of the Transvaal Museum. Ann. Transv. Mus. 8: 273-277. Camp, C. L. 1945. Prolacerta and the protorosaurian reptiles. Am. J. Sci. 243: 17-32, 84-101. CarRrROLL, R. L. 1975. Permo-Triassic ‘lizards’ from the Karroo. Palaeont. afr. 18: 71-87. CARROLL, R. L. 1976a. Eosuchians and the origin of archosaurs. Jn: CHURCHER, C. S., ed. Athlon: essays on palaeontology in honour of Loris Shano Russell. Misc. Publs. R. Ont. Mus.: 58-79. CARROLL, R. L. 1976b. Galesphyrus capensis, a younginid eosuchian from the Cistecephalus Zone of South Africa. Ann. S. Afr. Mus. 72: 59-68. CarRROLL, R. L. 1977. The origin of lizards. In: ANDREws, S. M., Mires, R. S., & WALKER, A. D., eds. Problems in vertebrate evolution. Linnean Soc. Symp. Ser. 4: 359-396. CarROLL, R. L. 1978. Permo-Triassic ‘lizards’ from the Karroo system. Part II. A gliding reptile from the Upper Permian of Madagascar. Palaeont. afr. 21: 143-159. Carroll, R. L. 1981. Plesiosaur ancestors from the Upper Permian of Madagascar. Phil. Trans. R. Soc. Lond. (B) 293: 315-383. Cott, H. B. 1961. Scientific results of an inquiry into the ecology and economic status of the Nile Crocodile (Crocodilus niloticus) in Uganda and Northern Rhodesia. Trans. zool. Soc. Lond. 29: 211-354. CurriE, P. J. 1978. The orthometric linear unit. J. Paleontol. 52: 964-971. CurriE, P. J. 1979. The osteology of haptodontine sphenacodonts (Reptilia: Pelycosauria). Palaeontographica. (A) 163: 130-168. Curri£E, P. J. 1980. A new younginid (Reptilia: Eosuchia) from the Upper Permian of Madagascar. Can. J. Earth Sci. 17: 500-511. CurriE, P. J. 1981. The vertebrae of Youngina (Reptilia: Eosuchia). Can. J. Earth Sci. 18: 815-818. Darsy, D. G. & OJACKANGAS, R. W. 1980. Gastroliths from an Upper Cretaceous Plesiosaur. J. Paleontol. 54: 548-556. GLADSTONE, R. J. & WAKELEY, C. P. G. 1932. The morphology of the sternum and its relation to the ribs. J. Anat. 66: 508-564. Gow, C. E. 1975. The morphology and relationships of Youngina capensis Broom and Prolacerta broomi Parrington. Paleont. afr. 51: 139-149. TANGASAURUS MENNELLI HAUGHTON 265 Harris, J. M. & Carrol, R. L. 1977. Kenyasaurus, a new eosuchian reptile from the Early Triassic of Kenya. J. Paleont. 51: 139-149. HaucGuton, S. H. 1924. On reptilian remains from the Karroo beds of East Africa. Quart. J. geol. Soc. Lond. 80: 1-11. HauGurTon, S. H. 1930. Notes on the Karroo Reptilia from Madagascar. Trans. R. Soc. S. Afr. 18: 125-136. HUENE, F. von. 1926. Gondwana Reptilien in Sudamerica. Palaeontol. Hung. 2: 72-73. HuENE, F. von. 1940. Die Saurier der Karroo-Gondwana und verwandten ablagerungen in fauitischer, biologischer und phylogenetischer Hinsicht. N. J. Min. Geol. Pal. 83: 246-347. HUuENE, F. von. 1952. Revision der Gattung Pleurosaurus auf Grund neuer und alter Funde. Palaeontographica (A) 101: 167-200. Kuan, O. 1969. Handbuch der palaoherpetologie, Teil 9. Proganosauria, Bolosauria, Placodon- tia, Araeoscelidia, Trilophosauridae, Weigeltisauria, Millerosauria, Rhynchocephalia, Pro- tosauria. Stuttgart: Gustav Fischer Verlag. McKiniay, A. C. M. 1956. Summary of the geology of Tanganyika. Part 1. Introduction and stratigraphy. Mem. geol. Surv. Tanganyika 1: 81-145. McKiniray, A. C. M. 1960. Brief comparison of the Karroo rocks of Madagascar and Tanganyika. Rec. geol. Surv. Tanganyika 8: 37-40. Noprcsa, F. 1924. On the systematic position of Tangasaurus and Saurosternon. S. Afr. J. Sci. 21: 206-207. Ortov, J. A. 1964. Osnovy Paleontologii. 12: 446-461. Moscow: Nauka. (In Russian.) Preyer, B. 1937. Die Triasfauna der Tessiner Kalkalpen. XII. Macronemus bassani Nopcsa. Schweiz. palaeont. Abh. 59: 3-140. PIVETEAU, J. 1925. Existence d’un Reptile a affinités lacertiliennes dans les formations permien- nes de Madagascar. C. r. hebd. Séanc. Acad. Sci., Paris 180: 154-157. PIVETEAU, J. 1926. Paléontologie de Madagascar XIII. Amphibiens et reptiles permiens. Annls Paléont. 15: 53-180. PIVETEAU, J. 1955. Eosuchia. In: PIvETEAU, J., ed. Traité de Paléontologie 5: 545-555. Paris: Masson. Reisz, R. 1977. Petrolacosaurus kansensis Lane, the oldest known diapsid reptile. Science N.Y. 196: 1091-1093. Romer, A. S. 1956. Osteology of the reptiles. Chicago: University of Chicago Press. Romer, A. S. 1966. Vertebrate paleontology. 3rd ed. Chicago: University of Chicago Press. Watson, D. M. S. 1957. On Millerosaurus and the early history of the sauropsid reptiles. Phil. Trans. R. Soc. Lond. (B) 240: 325-400. WILD, R. 1973. Die Triasfauna der Tessiner Kalkalpen, XXIII. Tanystropheus longobardicus (Bassani) (neue Ergebnisse). Schweiz. palaeont. Abh. 95: 1-162. ~ ’ _- = eat ~~. 7 x 7 — ~ - 9 ; ; a = b : ——. * ‘ 5 7 oe . i > ) ae ; ; _ } a a 7 ; 2 7 _ : - a : : -_ a : = f a a ye a PY. MP: 7 7 7 a oy = ya To a ae 7 7 A 6. SYSTEMATIC papers must conform to the /nternational code of zoological nomenclature (particularly Articles 22 and 51). Names of new taxa, combinations, synonyms, etc., when used for the first time, must be followed by the appropriate Latin (not English) abbreviation, e.g. gen. nov., sp. nov., comb. nov., syn. nov., etc. . An author’s name when cited must follow the name of the taxon without intervening punctuation and not be abbreviated; if the year is added, a comma must separate author’s name and year. The author’s name (and date, if cited) must be placed in parentheses if a species or subspecies is transferred from its original genus. The name of a subsequent user of a scientific name must be separated from the scientific name by a colon. Synonymy arrangement should be according to chronology of names, i.e. all published scientific names by which the species previously has been designated are listed in chronological order, with all references to that name following in chronological order, e.g.: Family Nuculanidae Nuculana (Lembulus) bicuspidata (Gould, 1845) Figs 14-15A Nucula (Leda) bicuspidata Gould, 1845: 37. Leda plicifera A. Adams, 1856: 50. Laeda bicuspidata Hanley, 1859: 118, pl. 228 (fig. 73). Sowerby, 1871: pl. 2 (fig. 8a—b). Nucula largillierti Philippi, 1861: 87. Leda bicuspidata: Nickles, 1950: 163, fig. 301; 1955: 110. Barnard, 1964: 234, figs 8-9. Note punctuation in the above example: comma separates author’s name and year semicolon separates more than one reference by the same author full stop separates references by different authors figures of plates are enclosed in parentheses to distinguish them from text-figures dash, not comma, separates consecutive numbers Synonymy arrangement according to chronology of bibliographic references, whereby the year is placed in front of each entry, and the synonym repeated in full for each entry, is not acceptable. In describing new species, One specimen must be designated as the holotype; other speci- mens mentioned in the original description are to be designated paratypes; additional material not regarded as paratypes should be listed separately. The complete data (registration number, depository, description of specimen, locality, collector, date) of the holotype and paratypes must be recorded, e.g.: Holotype SAM-—A13535 in the South African Museum, Cape Town. Adult female from mid-tide region, King’s Beach Port Elizabeth (33°51’S 25°39’E), collected by A. Smith, 15 January 1973. Note standard form of writing South African Museum registration numbers and date. 7. SPECIAL HOUSE RULES Capital initial letters (a) The Figures, Maps and Tables of the paper when referred to in the text e.g. °... the Figure depicting C. namacolus ...’; ‘. .. in C. namacolus (Fig. 10)...’ (b) The prefixes of prefixed surnames in all languages, when used in the text, if not preceded by initials or full names e.g. Du Toit but A.L.du Toit; Von Huene but F. von Huene (c) Scientific names, but not their vernacular derivatives e.g. Therocephalia, but therocephalian Punctuation should be loose, omitting all not strictly necessary Reference to the author should be expressed in the third person Roman numerals should be converted to arabic, except when forming part of the title of a book or article, such as ‘Revision of the Crustacea. Part VIII. The Amphipoda.’ Specific name must not stand alone, but be preceded by the generic name or its abbreviation to initial capital letter, provided the same generic name is used consecutively. Name of new genus or species is not to be included in the title: it should be included in the abstract, counter to Recommendation 23 of the Code, to meet the requirements of Biological Abstracts. PHILIP J. CURRIE THE OSTEOLOGY AND RELATIONSHIPS OF TANGASAURUS MENNELLI HAUGHTON (REPTILIA, EOSUCHIA) ALS s ha = im Sh ee WY ma ZZ. : ap) OSHLIWS Sagluvudil LIBRARIES SMITHSONIAN wn y e = a < <7 faa = oc jaa) Oo (oa) Dy Zz 4 ‘ INSTITUTION NOILALILSNI NVINOSHLIWS i z ay 6g, = w P| >) 2 > = > ae ae ae wD z wo > OSHLIWS Sa iyuvudii au B RAR I ES SMITHSONIAN = ; < = = Zz =| as ro) ae wn A OO © Se O = = zZ HSONIAN INSTITUTION NOILNLILSNI NVINOSHLIWS —s 4 (Tp) LR e Ww iE + aa + de aan a x = ac a O O Zz “y = LIBRARIES SSIaVESIT CLUB RAR TES. SMITHSONIAN _INSTITUTION NOILMLILSNI NVINOSHLIWS Ww SON “ = a SX si < 4 SSS : : 2 B + 5 a = XY PE: pa | Zz s SJIYVUGIT LIBRARIES SMITHSONIAN S oO z ‘s : : : = > : o — Pe ze # z INSTITUTION NOILALILSNI_ NVINOSHLINS aan Fd eo < < SS Vie <= ie 2. 7°) Foe = SS >" = Y) - ae ae Ww S3!1NVYGIT LIBRARIES SMITHSONIAN 2 ee J WW - Yyy “ - UD a > ae DW he : = fe 3 a Zz ad ; Zz we 5 om Zz ie S oO ea ee O a Se a w — his ES a SON be Rie 0} fae Pe) Fy Pe) x = : Ey : : 2 SS ra oe [= ae La at : 2) m 2 m oe m ie wn = wn = wn : HSONIAN INSTITUTION NOILALILSNI NVINOSHLINS S3IYWVYUEIT LIBRARIES SMITHSONIAN Save Ge Z 2 - Sk Ss = ‘zs = aN ee : = a : : g z a 2 g ae ®) : fF Zz = z fe eZ s > U = > : = faa Pa v”) mz se kOe 8 ee OSHLIWS $3 tuvydid LIBRARI ES SMITHSONIAN _INSTITUTION NOUN = —~| UT ea ams “4 ios — = se = PY = z a G&G fara CZ Pe ao = : : 2 Ge 3 : : See Z els g 2 ee “ HSONIAN” INSTITUTION | NOILALILSNI NWINOSHLIWS >? l¥vud Be LIBRARI Ss yy vere Oo jaan S oO as Oo oY by ioe) — w ‘ — w = Gy, > : 2 Ye 5 2 : Yghir y S S , lp f= = 2 WWE > : en ee ad = aN \ — 3 m Y) faa S\N Ww) m Ww) > s 2 z w = ” = w” = JOSHLIWS S43 \Yuvudl me B RAR! ES SMITHSONIAN INSTITUTION NOILALILSING Bs z , Ee < j < wy < = z ad = My, = 2 iby, O ae RN O oO Vp, 4 Y,, BE & Z Sy 2 Zeya > = a > >” = Fa 7) 2 2 Zz wn HSONIAN INSTITUTION NOILALILSNI NLILSNI NALILSNI ARLES LIBRARIES SMITHSONIAN RARIES SMITHSONIAN ALILSNI RARIES se Pa fF Y UP O> SER <_ eS a: a eS =< +t ay fe te he X = “llffd = il = ak = Vas = ial > UgiT_ LIBRARIES SMITHSONIAN INSTITUTION NOILOLILSNI_NVINOSHLINS Saluvuaiy_b & in a oi a us .o us a us z = oc = or ee. ox = z < = < = < a | : : ; : : : 3 z ay s z a 2 eka Pad UTION > NOWLNLILSNI NVINOSHLIWS ~°4 l\YVvVugd fae LIBRARI oe ideel LEI hy mi O i) fo) ~ Oo y = re) — wm > ae w = w ’ > : = > 2 Ce a = > 7 NOILALILSNI_NVINOSHLINS _LIBRAR KES _INSTITUTION ed, “ 7 Ww < it = us faa) = m = or — co 7 oO pia (@) aS oO ae Udit LIBRARIES SMITHSONIAN INSTITUTION NOILOALILSNI NVINOSHLINS S3JIYVYEIT LY 5 Z is z is = - wo 2 re) 2 ow 2 ob Pe = » 5 re) Ss 20 > — egy > E oa = > 5) = “7 “iy, ee] i= > 6 i Po} | 9 “Yh im D = D = wn = 7) = 7) = wn UTION NOILNLILSNI NVINOSHLINS SSIYVYEIT LIBRARIES INSTITUTION N n z w z ; wm z oe: w” \ = <= = ee = < \ . = “os S oh a. = 5B OS Bey D a, A D n \ w , vs = = = E = = =e > = > ‘ = > = > See _ z= Ww Ea Nyls Yn ee Yi SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLINS S3I1YVYUdIT_ L Z i Mies se z eu a Z ” «” aay w n a o a oc et a za 4 = NY >" = > = = ” bisee 2 77) Zz 3 nn _ NVINOSHLIWS SJ'IYVYEIT LIBRARIES INSTITUTION Ni = ” S wo us a oN nee a KE wor aes a = AS x L 2 We = (SM) & ey =< WSs SNE ee Dy; Peters ene ee sy por b nent eters SPN Ait ot De snne eA Coan NT hbraren wrt 4.0 Gath GaN rations gee as LEPC VERE teas tee th 2 Oy ne stag « ‘ wv SMITHSONIAN INSTITUTION LIBRARIES Ml t oa ‘ ee un ‘ 1 S Teal (ica ie Ot Se a LU 3 9088 01206 6783